Avenue des Nerviens 9-31 - 1040 Brussels - BELGIUM - Tel. +32 2 514 11 11
[email protected] - www.fooddrinkeurope.eu - ETI Register 75818824519-45
Copyright FoodDrinkEurope aisbl
Precautionary Allergen
Labelling (PAL): a science-
based approach based on
Quantitative Risk
Assessment
FoodDrinkEurope non-paper
Version 2, January 2021
2/27
Abstract/Executive summary
Allergens as ingredients have been regulated in the EU since 2005. Due to the realities of food
production there are situations where it is not possible to avoid the unintended presence of
allergens, which in some cases might pose a risk to susceptible people. Precautionary allergen
labelling (PAL) has evolved as an essential tool both to communicate and manage this risk.
Mounting evidence indicates that PAL is increasingly losing credibility among stakeholders,
including critically those for whom it is primarily intended, i.e. allergic consumers. This failure of PAL
diminishes a valuable and necessary risk management tool and places allergic consumers at risk.
The current issues associated with PAL can be attributed to several factors, in particular the
absence of generally agreed quantitative limits for the unintended presence of allergens. The lack of
harmonised standards across Member States and industry also adds to confusion.
Currently PAL is not formally regulated in the EU, although the general principles of food safety law
arguably apply to it. However, Article 36 of Regulation (EU) 1169/2011 (hereafter: the “FIC
Regulation”) sets out a framework which can be used to implement a comprehensive, consistent,
and science-based approach.
New methodologies as well as growing volumes of good quality data now exist that can be applied
to assess the risk from unintended allergen presence.
This paper describes how adoption of quantitative limits (based on reference doses) could form part
of an EU-wide approach aligned with the principles enunciated in Article 36 of the FIC Regulation.
Such an approach would greatly strengthen the protection of allergic consumers by making PAL
meaningful and transparent. It would also benefit the European food industry by providing a
comprehensive framework (“level playing field”), which would also strengthen the single market.
FoodDrinkEurope would like to see a defined framework for the application of PAL which meets the
requirements of article 36(2) of the FIC Regulation, namely:
- PAL should be clear: a single statement with a single meaning, easy to translate into EU
languages, i.e. “may contain [allergen]”.
- PAL should not be misleading: it should only be applied where a defined, appreciable
risk has been identified, including (where it is relevant and possible) through a
quantitative risk assessment.
- PAL should be applied based on transparent quantitative limits derived using the most up to
date, relevant, peer-reviewed and robust scientific data.
- Consumers need to know that products have been through a risk assessment and that the
presence or absence of PAL is a consequence of that process.
3/27
Summary Recommendations on PAL
FoodDrinkEurope would like to see a defined framework for the application of PAL which meets the
requirements of Article 36(2) of the FIC Regulation, namely:
• Clear: a single statement with a single meaning, easy to translate into EU languages, i.e. “may
contain [allergen]”
• PAL should not be misleading: it should only be applied where a defined, appreciable
risk has been identified, including (where it is relevant and possible) through a quantitative risk
assessment.
• PAL should be applied based on transparent quantitative limits derived using the most up to
date, relevant, peer-reviewed, and robust scientific data.
• When Quantitative Risk Assessment (QRA) relies on analytical data, it should be noted that
sampling procedures and analytical methods have limitations regarding sensitivity, specificity,
and accuracy. The applicability of analytical data as an input into QRA requires harmonization.
• In addition, consumers need to know that products have been through a risk assessment and
that the presence or absence of PAL is a consequence of that process.
4/27
Contents
1. Introduction .......................................................................................................................... 5
2. Definition and purpose of PAL ............................................................................................ 6
Definition ....................................................................................................................................... 6
Purpose ......................................................................................................................................... 6
Who are we trying to protect and against what? The challenges of making PAL meaningful. ........ 6
Why do we need PAL to protect allergic consumers? ........................................................ 8
Current situation in the EU (and beyond) - why isn’t PAL working? ............................................... 8
3. Legislative provisions of relevance for PAL .................................................................... 10
Regulation (EU) 1169/2011 (the FIC Regulation) ............................................................ 10
Regulation (EU) 178/2002 (the General Food Law) ..................................................................... 11
4. The scientific basis for PAL .............................................................................................. 12
Why does the application of PAL have to be based on sound science? ...................................... 12
Allergen hazard characterisation and risk assessment ................................................................ 12
The need for harmonised quantitative limits (zero tolerance is not zero risk) ............................... 15
The role and limitations of analytical methodologies ........................................................ 15
Further considerations in the application of PAL .......................................................................... 16
Consumer and health care professionals (HCP) perspectives on PAL and relevance to risk ....... 16
The twin roles of PAL: risk communication and risk management ................................... 17
5. Implementation of PAL: towards more consistent allergen risk communication .......... 19
Which substances and products should be covered by PAL? ...................................................... 19
Should PAL be clearly distinguished from the allergen information given in the list of ingredients?
.................................................................................................................................................... 19
Expression and presentation of PAL ........................................................................................... 19
i) Wording for precautionary allergen statement .......................................................... 19
ii) Location of wording .................................................................................................. 20
iii) Other legibility aspects .......................................................................................... 20
• General principles: ................................................................................................... 20
• Font size: ................................................................................................................. 20
• Emphasis: ................................................................................................................ 21
PAL for non-prepacked foods ...................................................................................................... 21
Annex I ........................................................................................................................................... 27
5/27
1. Introduction
Allergens are common constituents of consumer products with valuable functional and/or nutritional
attributes. Food allergy refers to an inappropriate immune response to a food constituent (almost
always a protein), causing the food to provoke an allergic reaction when it is eaten again. The
nature and type of reactions, as well as the numbers of people with food allergies make allergens
an important food safety issue.
Allergen incidents can arise for many reasons along the food chain, including unintended presence
of allergens which are not part of the recipe. Despite the most serious efforts to manage allergens
during manufacturing and other operations, this unintended presence of small amounts of allergens,
which can pose a risk to susceptible individuals, cannot always be avoided. Precautionary allergen
labelling (PAL) was introduced as one of the measures to mitigate and manage this risk.
While current practices in the management of major allergens (according to Annex II of the FIC
Regulation) have increased the safety of food products for allergic consumers, the lack of an
agreed, consistent approach to quantitative risk assessment for unintended allergen presence has
led to divergent standards applied by different manufacturers, as well as divergent approaches by
enforcement authorities across Europe and the world. These factors have led to the extensive use
of PAL, that may not be related to the actual risk the product poses, does not always cover the right
allergens and restricts the food choices of allergic consumers while damaging its credibility. This, in
turn has led to a significant proportion of these consumers taking risks and to allergic consumers
suffering accidental reactions, as documented in various publications (Barnett et al. 2011, Blom et
al. 2018, Cochrane et al. 2013, DunnGalvin et al. 2015, Michelsen-Huisman et al. 2018).
PAL remains a necessary and useful tool to manage and communicate risk to allergic consumers
but, in order to restore its value and maximise consumer protection, an urgent need exists for the
adoption of a comprehensive EU approach to quantitative risk assessment for PAL purposes.
The European Food Safety Authority addressed the issue of food allergens in an Opinion in 2014
(EFSA 2014). While the Opinion discussed risk assessment approaches, it was limited only to
(regulated) allergens used as ingredients. Thus, it did not consider how the risks arising from
unintended allergen presence might be assessed, which would have provided a sound scientific
basis for risk managers to implement PAL.
The FIC Regulation requires implementing acts to be adopted by the Commission concerning rules
for the use of PAL. These include that the information provided must not be misleading, must be
clear and that it must be based on relevant scientific data. In this paper, FoodDrinkEurope proposes
an approach to PAL based on quantitative risk assessment, guided by the principles set out in the
Regulation
1
. The approach is aimed at the application of PAL to pre-packaged retail food products
for normal consumption, although elements could be applied to other sectors.
1
The considerations on risk assessment included in this paper apply to substances or products causing allergies. Care
needs to be taken to differentiate food allergy from food intolerance, which does not involve the immune system. However,
the “PAL statements” are currently used also to inform consumers about the possible presence of substances or products
causing intolerances. Therefore, the considerations on risk communication included in the section “Implementation of PAL:
towards more consistent allergen risk communication” also apply to the communication about substances or products
causing intolerances.
6/27
2. Definition and purpose of PAL
Definition
Precautionary allergen labelling (PAL), also called advisory labelling, refers to the voluntary labelling
to indicate that one or more regulated allergens could be unintentionally, but unavoidably, present in
a product, and thus pose a risk to susceptible consumers. Currently, there are no formal legal
definitions of Precautionary Allergen Labelling in the EU, nor a framework for its application,
although the FIC Regulation details a possible basis for such a framework in Article 36, as
discussed later. Guidance on good practice for its application has been published, for example by
the UK Food Standards Agency and FoodDrinkEurope.
Purpose
PAL serves both to communicate risk, but also manage it, its ultimate purpose being to avoid
reactions to allergens in susceptible consumers. The terminology used for PAL aims to convey to
susceptible consumers the possibility that an allergen may be present in a product and therefore
pose an appreciable risk to them, which the manufacturer wishes them to avoid. In other words, a
food producer should be using PAL primarily to dissuade susceptible consumers from consuming
their product.
Unintended allergen presence can occur in a number of ways, the most common and best known
being through cross-contact during manufacture of either the product or one of its components,
including agricultural raw materials. However, situations that can give rise to unintended allergen
presence encompass the whole supply chain from the fields in which agricultural commodities are
grown through the containers in which those commodities are transported, right up to storage at the
manufacturing location. Unintended allergen presence can also manifest itself in several different
ways depending on the production process, the source of unintended presence and the physical
form of the allergen (readily dispersible or particulate). One situation can be a very low level of
allergen present in all units of the product. Another can include presence of the allergen in a
proportion of units only, due to carryover at changeover. In cases of allergen in particulate form,
most units may not have any allergen, but where it occurs, it may be sufficient to provoke a severe
reaction, so resulting in a rare event, but with serious consequences.
Inherent in the concept of PAL is a degree of uncertainty about the actuality of the risk (the allergen
may or may not be present). There is also uncertainty about the exact nature of the risk to which
allergic consumers may be exposed, such as likelihood of and severity of reaction if allergen is
present, in part because of variability in their susceptibility, but also because currently, no
quantitative limits have been agreed which are generally accepted by authorities, including national
and Union authorities in the EU.
Precautionary allergen labelling is not an appropriate strategy to manage lack of Hazard Analysis
and Critical Control Points (HACCP) during manufacturing, or lack of adherence to generally
recognised Good Manufacturing Practices (GMP).
Who are we trying to protect and against what? The challenges of making PAL meaningful.
People with allergies vary over a very wide range in the minimum dose that will elicit a reaction, with
data from controlled food challenges pointing to a million-fold range (micrograms to grams).
7/27
Manifestations of allergic reactions also vary considerably, from those barely perceptible to the
affected person and not evident to an external observer (subjective reactions) through objective
reactions of various degrees of severity to life-threatening anaphylaxis involving compromise to the
cardiovascular and respiratory systems. These observations highlight the challenges in defining
limits, based upon reference doses as the basis for the transparent and consistent use of PAL.
Using PAL in a meaningful way for consumers is not simply about setting limits so low that every
allergic consumer is protected against every possible reaction, however mild, but being realistic that
excessive use of PAL, which would ensue, in fact undermines its purpose in minimising the number
of reactions among susceptible consumers. The challenge is therefore to strike the right balance
between reference doses that are highly protective of very reactive consumers, while ensuring that
they do not result in such proliferation of PAL use that its credibility is undermined, allergic
consumers are driven towards risk-taking behaviours, and the overall risk to them is in practice
driven up (Figure 1).
Figure 1: conceptual representation of the relationship between the extent to which PAL is
used and how well it is observed. The blue line shows how the expected number of reactions
varies with the reference dose (a lower reference dose protects a higher proportion of the
population). The green line shows the extent (%) to which PAL is observed according to the extent
of its use. Where a high proportion of products bears a PAL statement, choice for allergic
consumers, as well as credibility of PAL is reduced and so is observance.
PAL would therefore serve its optimal risk management purpose, if it were applied based upon
reference doses that avoid the occurrence of reactions harmful to health by ensuring its
circumspect, transparent and judicious use.
On a population basis, this judicious application of PAL would ultimately be most protective. It
cannot avoid the occurrence of every mild reaction, but would be protective against life-threatening
reactions, all by re-establishing the credibility of PAL and therefore offering a valuable and
trustworthy risk communication tool to producers and consumers. A public health goal for PAL
should be to minimise the number of reactions in the susceptible population. This requires
8/27
consideration not only of the limits that would be set as a basis for deciding whether or not PAL
should be used, but also to socio-cultural factors such as the credibility of PAL, which is critical to its
effectiveness and closely linked to the extent of its use, as well as other factors discussed above.
Why do we need PAL to protect allergic consumers?
Food production operates under a range of technical, scientific, legal and economic constraints
which bear upon the need to use PAL.
From a technical perspective, the extreme diversity of food products and equipment, as well as
conflicting food safety requirements often impose the need for PAL. Thus, equipment may not be
designed to meet evolving food safety or allergen considerations, hence the importance of HACCP
and GMPs. Operating it to meet current standards, particularly in the absence of any “official”
definition of such standards can therefore be extremely challenging (Stone and Yeung 2010).
Effective means of removing allergens, such as wet cleaning are either impractical in some
circumstances and could indeed lead to microbiological safety issues (e.g. dry mix operations)
(Jackson et al. 2008, Röder et al. 2010). Effective means of controlling allergen carry-over, such as
scheduling are also subject to other limitations, such as taste, colour, etc. Other conflicting
requirements include minimisation of waste and more generally environmental impact.
Legal constraints can also impinge on the need to use PAL. For example, grain standards, which
are the basis of world trade, allow for the presence of other grains than the one nominally sold (e.g.
soy in wheat) in proportions that are potentially significant from an allergen perspective. However,
the use of PAL does not exonerate the manufacturer from potential liability, if it is not accompanied
by evidence of adherence to good manufacturing practices and an allergen management plan.
Economic constraints and those related to environmental impact also cannot be ignored. It is
recognised that dedicated equipment and lines offer an effective solution but are only practicable in
a limited number of situations and usually for one or two allergens, rather than all regulated
allergens. Cleaning protocols are also limited not only by the resources needed, but by the
downtime they impose. The absence of defined standards, applicable to all, may impose a potential
competitive disadvantage in favour of less stringent protocols.
Current situation in the EU (and beyond) - why isn’t PAL working?
There is a growing volume of evidence that PAL as currently used in the EU, as well as elsewhere,
fails to achieve its goal of protecting vulnerable consumers (Allen, Turner et al. 2014, Blom et al.
2018, Dunn Galvin et al. 2015, Michelsen-Huisman et al. 2018). This failure manifests itself in the
extent to which it is disregarded by allergic consumers (Barnett et al. 2011, Cochrane et al. 2013,
Soon & Manning 2017), which has the practical result of putting them at risk.
Research among allergic consumers reveals several reasons why PAL is failing, such as:
- proliferation of PAL in certain product categories;
- confusing terminology giving the impression of a risk hierarchy, unsupported by experimental
evidence;
- lack of transparency over its use;
- lack of understanding of the framework in which PAL is used (e.g. an assumption that it is
mandatory);
- inappropriate use, e.g. on products where it is unexpected;
- lack of agreed standards for application.
9/27
Clearly, some of these factors are very closely related to and influence each other and this is
covered in detail in Madsen et al 2020. All these factors have stimulated a significant proportion of
allergic consumers to undertake their own risk assessments, although without a sound evidence
base, thereby putting themselves potentially at risk.
The efficacy of PAL is thus affected by the circumstances, reasons for and extent of its use, all of
which affect consumers’ perception of PAL and therefore their trust. Thus, the consequences of
using PAL are not limited to what is communicated to the consumer (discussed later). Circumspect
and responsible use of PAL is critical to successfully achieving its goal, as illustrated conceptually
for the interrelationship between reference dose, the extent of the use of PAL and how well allergic
consumers adhere to the warnings in Figure 1 above.
Enforcement authorities’ attitudes to the use of PAL in the EU as well as beyond vary across
countries. Some authorities consider that the presence of any detectable allergen which is not an
ingredient, using any analytical technique, infringes the Food Safety Law (Regulation 178/2000)
unless a PAL is applied. This zero-tolerance approach inevitably leads to ever-increasing numbers
of products bearing PAL, as the sensitivity of analytical techniques continues to increase.
Other authorities use quantitative risk assessment to determine whether a product warrants a PAL
statement, even though allergen might be detectable, in line with their own guidance to industry with
regard to the application of PAL. However, there remains a lack of transparency concerning the
limits which they use and how they take account of different variables in coming to a decision. It is
also unclear whether, when using this approach, authorities have developed a common
methodology. The diversity in PAL management decisions from different countries has been
summarised in Madsen et al. (2020).
10/27
3. Legislative provisions of relevance for PAL
Regulation (EU) 1169/2011 (the FIC Regulation)
In order to enable consumers, particularly those suffering from a food allergy or intolerance, to make
informed choices which are safe for them, the FIC Regulation makes it mandatory to provide
information on the presence of these substances in foods and drinks (Article 9.1 (c)). Box I provides
more details on this.
The FIC Regulation refers to PAL for the very first time and provides a legal basis for adopting EU
rules in this respect. More precisely, Article 36(2) clarifies the requirements applying to voluntary
food information (and, thus, also to PAL): this must not mislead the consumer, it must not be
ambiguous or confusing for the consumer; and it must, where appropriate, be based on the relevant
scientific data.
Furthermore, Article 36(3)(a) of Regulation (EU) 1169/2011 provides a legal basis for rules on
voluntary information on the possible and unavoidable presence in food, due to cross-
contamination, of substances causing allergies or intolerances (PAL). No deadline for adopting this
act is set by the Regulation. As mentioned before, there are thus far no formal legal definitions of
PAL in the EU.
11/27
Regulation (EU) 178/2002 (the General Food Law)
The use of PAL has been historically based on the principles laid down in Regulation (EU) 178/2002
(the General Food Law). Article 5 of this Regulation states that food law must pursue, among
others, a high level of protection of human life and health and the protection of consumers' interests.
In order to achieve this objective, when appropriate, food law shall be based on risk analysis. Risk
assessment shall be based on the available scientific evidence and undertaken in an independent,
objective and transparent manner. Risk management shall take into account the results of risk
assessment (Article 6).
Article 14 of the General Food Law refers to the general principle that unsafe food cannot be placed
on the market. Thus, a legal obligation exists for food business operators to ensure that the food
which is offered for sale to consumers is safe. In determining whether any food is unsafe, regard
shall be given to the information provided to the consumer on the adverse health effects that the
food can have, also taking into account the particular health sensitivities of a specific category of
consumers (as they may be allergic or intolerant consumers). PAL is therefore relevant in order to
ensure that safe food is offered to consumers and achieve the high level of consumer protection
required by the Regulation.
Lastly, Article 16 of the General Food Law and Article 7 of the FIC Regulation provide that food
information cannot be misleading for consumers; on the contrary, this must be accurate, clear and
easy to understand for consumers.
Avenue des Nerviens 9-31 - 1040 Brussels - BELGIUM - Tel. +32 2 514 11 11
[email protected] - www.fooddrinkeurope.eu - ETI Register 75818824519-45
Copyright FoodDrinkEurope aisbl
4. The scientific basis for PAL
Why does the application of PAL have to be based on sound science?
PAL aims to convey to a vulnerable subpopulation that a hazard could be present in a food
such as to pose a risk to some of that subpopulation. As discussed above, the efficacy and
therefore value of PAL relies critically on its credibility and the lack of an agreed sound
scientific basis to its current use undermines its value and the protection it can afford allergic
consumers. PAL is about risk management and risk communication. Since PAL aims to
convey a risk, its application should follow a thorough risk assessment which should be
quantitative whenever possible.
The FIC Regulation highlights the importance of sound science as the basis for PAL: Article
36(2)(c) states that “it shall, where appropriate, be based on the relevant scientific data”. The
last 10-20 years have seen considerable progress in the development of risk assessment
approaches for allergens, as well as in the generation of data to do these risk assessments,
as recognised in the EFSA Opinion (EFSA 2014) and has been built upon since then. These
data and knowledge can thus provide the sound scientific basis for PAL and the means to
perform quantitative risk assessments and propose quantitative limits (based on reference
doses) for its application (Allen et al. 2014, Blom et al. 2019, Crevel et al. 2014, DunnGalvin
et al 2019, Madsen et al. 2020, Remington et al. 2020, Taylor et al. 2014, Westerhout et al.
2019, Wheeler et al. 2019 ).
Allergen hazard characterisation and risk assessment
Hazard characterisation is one of the cornerstones of risk assessment and it is useful to
review it briefly in the context of allergens.
Hazard characterisation for allergens has advantages compared to characterisation of both
chemical and microbiological hazards, as it relies on human data. Animal to man
extrapolation of the results of toxicological studies and consideration of various other
qualitative and quantitative uncertainties associated with non-human toxicity data are
unnecessary.
The need for human data also imposes ethical and practical constraints that limit both the
amount and type of data that can be generated. Challenge studies rely on volunteers who
can only be tested a limited number of times and may not be fully representative of the
whole population allergic to a food. The availability of suitable clinics and trained personnel,
as well as the prevalence of allergy to a particular food further limit the numbers that can be
tested. All these factors delayed the availability of good quality data in adequate quantities.
Increasing amounts of data of ever better quality continue to become available on the
relationship between minimum eliciting dose and frequency of reaction in the population
allergic to a number of priority allergens. These data also demonstrate that minimum doses
for the elicitation of allergic effects by food allergens exist at an individual level and thus also
at a population level. Ideally, everyone at risk of reacting would be protected by setting the
*The Eliciting Dose (ED), where EDp refers to the dose of total allergen protein predicted to produce a
response in p% of the allergic population, represents the dose of an allergen (EDp) at which a
proportion of the allergic population would be likely to react to but, importantly, does not identify a
dose below which no allergic individual would react. Thus, the ED
01
and ED
05
are the doses at which
only 1% and 5%, respectively, of the allergic population would react with objective symptoms.
limits above which PAL must be used lower than the lowest Minimum Eliciting Dose (MED)
in the population. However, such an approach is currently unfeasible because the lowest
MED has not been determined for any allergen, and furthermore, limits derived in such a
way would probably be unattainable in most current food production practice. This would
result in PAL being used on most products, in circumstances where the risk to the
overwhelming majority was negligible. A more practicable solution would be to use an
approach such as that proposed by the VITAL Scientific Expert Panel. Using available data
and statistical modelling techniques, an international scientific expert panel set up as part of
the Australia-New Zealand Voluntary Incidental Trace Allergen Labelling (VITAL) initiative
proposed Reference Doses (RDs) for most of the European regulated allergens. VITAL 2.0
RDs were released in 2012 and the science underpinning them published in peer-reviewed
journals in 2014 (Allen et al. 2014, Crevel et al. 2014, Taylor et al. 2014). The RDs resulted
from a joint effort by TNO (Netherlands) and FARRP (US), facilitated by the VITAL Scientific
Expert Panel (VSEP), to model human eliciting dose (ED)* data for a range of food
allergens. Where the data were of sufficient quality and quantity, the ED
01
was used as the
basis for the RD. Where the data were insufficient to allow estimation of the ED
01
, the lower
95% confidence interval of ED
05
was used. The authors recognised that a very small minority
might still be at risk of more significant reactions, although they would still benefit, and
emphasised the need to communicate this clearly.
Whilst the VITAL approach and VITAL 2.0 RDs have been well received by stakeholders,
and many food companies and authorities (ANSES, FSA) have endorsed their use for risk
management purposes or considered them for more general enforcement purposes
(Germany - Waiblinger & Schulze, 2018; Belgium - SciCom 2017), there has been a
continued lack of consensus regarding regulation of PAL by authorities. Therefore, to further
support and develop the use of RDs, further research has been undertaken that has
generated additional data and new methodologies, including a stacked modelling averaging
approach, resulting in the publication of updated population minimum eliciting dose
distributions for use in risk assessment and release of VITAL 3.0 RDs (Allergen Bureau
2019, Houben et al 2020, Remington et al 2020, Westerhout et al 2019, Wheeler et al 2019,
2020).
From a scientific perspective, the current use of PAL reflects uncertainty about both the
extent and nature of the risk posed by allergens, which can result in a hazard-based
approach to risk management (i.e. if there is a possibility that allergen may be present, PAL
is always used). However, the extent of the risk, in terms of the populations at risk and the
distribution of MEDs (individual minimum eliciting doses) in those populations is now well
understood, as discussed above. The nature of the risk, in terms of the type of reaction
provoked by a defined dose of allergen has been more uncertain, although evidence
indicated lower doses were associated with a lower probability of severe reactions (Rolinck-
Werninghaus et al 2012, Crevel et al 2014). Research into the nature of reactions at different
doses is emerging to support this. For example, in the Peanut Allergen Threshold Study
(PATS) only 8/375 (2.1%) of subjects had a convincing objective reaction to the VITAL 2.0
ED
05
for peanut and all were considered to be mild (Hourihane et al (2017). Additionally, the
TRACE (Threshold Reactivity Clinical Evaluation) study has provided information on how
individual minimum eliciting doses in peanut allergic adults are altered by co-factors (sleep
deprivation and exercise) indicating that the proposed VITAL RD would not need adapting to
account for this (Dua et al., 2019).
Madsen et al. 2020 have reviewed and summarised the scientific progress in this area
concluding that sufficient knowledge exists to implement a proposed framework for reaching
consensus on a defined level of protection for allergic consumers, such that PAL can be
*The Eliciting Dose (ED), where EDp refers to the dose of total allergen protein predicted to produce a
response in p% of the allergic population, represents the dose of an allergen (EDp) at which a
proportion of the allergic population would be likely to react to but, importantly, does not identify a
dose below which no allergic individual would react. Thus, the ED
01
and ED
05
are the doses at which
only 1% and 5%, respectively, of the allergic population would react with objective symptoms.
applied on the basis of transparent quantitative limits (based on reference doses). Madsen et
al. hoped to trigger cross-stakeholder engagement and collaboration to define appropriate
levels of protection for food-allergic consumers, calling upon Competent Authorities to
champion and lead the activity, which FoodDrinkEurope also support.
Avenue des Nerviens 9-31 - 1040 Brussels - BELGIUM - Tel. +32 2 514 11 11
[email protected] - www.fooddrinkeurope.eu - ETI Register 75818824519-45
Copyright FoodDrinkEurope aisbl
Where food consumption data are available, additional assessments based on probabilistic
modelling techniques can be used, which take account of the uncertainty and variability
associated with each input variable. Because their output is a range of values generated
from probabilistic distribution functions, they also negate the need to apply, often arbitrary,
uncertainty factors to the risk assessment output (e.g. Rimbaud et al. 2010).
The need for harmonised quantitative limits (zero tolerance is not zero risk)
When PAL was first introduced quantitative data on minimum eliciting doses, as well as on
the size of the at-risk population hardly existed. Risk assessment was therefore extremely
challenging in the face of this high degree of uncertainty. Not surprisingly, as there was little
basis other than analytical limits of detection to base quantitative limits, those were often
adopted by authorities as well as risk managers in industry in an approach designated “zero
tolerance”. However, “zero tolerance” obviously cannot be equated with zero risk.
An even more serious consequence of the high level of uncertainty and lack of quantitative
standards was that application of PAL lacked consistency across industry, with decisions
effectively based on each risk manager’s understanding and perception of the risk.
Therefore, every food manufacturer’s PAL statement potentially referred to a different level
of risk/level of protection, limiting the capacity of PAL to communicate effectively risk to
consumers, with the consequences already described earlier.
A prerequisite to restoring the credibility and value of PAL is therefore to assure consistency
of its application, such that a PAL statement indicates a defined level of risk. Quantitative
limits are a critical element in defining such levels of risk at a population level. The data and
techniques have already been developed and indeed deployed in initiatives such as the
VITAL Program. Clearly, a zero-tolerance approach cannot provide this assurance because
of the variability in the capacity of assays to detect and quantify allergens of interest,
particularly given the extreme diversity of food matrices. Furthermore, it is based on
detection limits rather than dose, which is acknowledged to be the appropriate metric. More
fundamentally, a zero-tolerance approach presents ever-changing standards, and therefore
works against transparency and the establishment of generally agreed common standards,
which are critical to restoring the credibility of PAL.
The role and limitations of analytical methodologies
A necessary prerequisite of the practical application of PAL for all stakeholders, particularly
producers and authorities, is the availability of reliable and practical analytical methods.
Producers need to validate analytically and subsequently verify their allergen management
measures. Authorities rely on analysis to verify whether products conform to provisions on
food safety and information regarding allergens.
Currently, the most commonly used analytical technologies are based on antibody
recognition (ELISA – Enzyme-Linked Immunosorbent Assay) and DNA sequencing (PCR –
Polymerase chain reaction. Both these technologies have limitations with regard to
sensitivity, accuracy and specificity, which makes their deployment challenging. Analytical
results can vary significantly, depending on the methods, equipment and/or test-kits used,
the product matrices as well as the laboratories operating the analysis. The lower the values
analysed, the greater the possible error rates.
The scientific analytical community recognised these issues as attested by publications such
as Johnson et al. (2014) from the EuroPrevall consortium. EFSA also acknowledged these
issues in its 2014 Opinion. While these issues affect both zero tolerance and risk-based
approaches, to verification and enforcement, they are particularly pertinent to quantitative
approaches. One particular aspect, but by no means the only one, noted by both the
EuroPrevall consortium and EFSA was the lack of certified reference material and methods
for the effective quantification of all allergens as listed in Annex II of EU Reg 1169/2011. In
response there has been significant work undertaken and progress made in addressing
these issues, though challenges do still remain (Holzhauser et al. 2020).
Effective implementation of quantitative limits (based on reference doses) will require
development and implementation of methods and protocols capable of reliably and
accurately detecting allergens at relevant concentrations
Further considerations in the application of PAL
Protection of allergic individuals is a shared responsibility among all stakeholders, which
necessitates a clear communication about what the use of PAL implies. In particular, it
needs to be clear that its application is aimed at foods for normal consumption, with the
implication that there may be a very small number who cannot be protected totally against
reactions given current knowledge and practice. Risk assessment must therefore also take
account of stakeholder understanding of PAL and behaviours in relation to it.
Consumer and health care professionals (HCP) perspectives on PAL and relevance to
risk
Stakeholder perspectives on PAL, including those of allergic consumers and HCPs, were
described as part of the iFAAM project. For consumers, the use of PAL is seen as
inconsistent and lacking transparency, not helped by misunderstanding about its legal status
(voluntary vs mandatory). This perception is not helped by a poor understanding of PAL
statements by many consumers and some healthcare professionals (HCPs) who advise
them. Consumers with food allergies respond by being very selective in their food
purchases, which gives rise to extra costs, anxiety, and impaired quality of life. The
proliferation of PAL, together with its appearance on unexpected products, has led to a loss
of credibility (Barnett et al. 2011, DunnGalvin et al. 2015, 2019a, 2019b, Soon & Manning
2017) and reduced observance by consumers. There is a clear relationship between the
extent to which PAL is used and the extent to which it is observed by consumers. In
December 2019 the European Federation of Allergy and Airways Diseases Patients’
Associations (EFA) published the final report of the Food DETECTives project focusing on
the quality of life of people with food allergies in Europe. Within this report a key
recommendation for regulators is to establish RDs and a harmonized quantitative risk-based
approach to applying PAL.
The attitudes of HCPs have not been as extensively studied as those of consumers with
food allergies. Nevertheless, they often mirror those of allergic consumers in terms of their
interpretation of PAL statements as a hierarchy of risk. Less than 60% HCPs recommended
total avoidance of products with PAL, a message at variance with the intent of industry risk
managers (Turner et al. 2014). Those working as allergy specialists were even less likely to
recommend stringent avoidance, possibly because of their awareness of the impact of
limited choices on their patients’ lives or because of lack of knowledge of voluntary industry
systems such as status of PAL (Turner et al. 2015). Avoidance advice was differentiated
according to medical history, with more stringent avoidance advised for those with co-
existent asthma, prior anaphylaxis or previous reaction to a tiny amount of allergen.
The twin roles of PAL: risk communication and risk management
PAL has two closely linked roles: risk communication and risk management. The
communication element is to inform at-risk consumers that the product in question could
precipitate a reaction. It is critical that those consumers understand the meaning of the
warning, rather attempt to do their own risk assessment, for which they do not have the right
information. It therefore also is incumbent on manufacturers and suppliers of products to
understand how consumers interpret such warnings, as well as to be clear about how they
want their message to be interpreted. The message should thus be clear and indicate that
the food producer’s considered judgement, based on a risk assessment, is that the product
is not suitable for people with the relevant allergies. The basis of that judgement should also
be clear, hence the need for agreed, consistent limits, as discussed in the preceding
paragraph.
While clarity and transparency are critical to PAL, they are also required for the converse
situation, i.e. when PAL is not used, as already discussed. Thus currently where no PAL
statement is present, this may mean one of two things: (1) the manufacturer has performed
a risk assessment and deemed the product not to require PAL because the risk is negligible
(because the allergen content per portion is below the reference dose) or (2) the
manufacturer has not done a risk assessment, e.g. owing to lack of understanding,
knowledge or awareness of unintended allergens. Under current circumstances, where there
is no common standard for the application of PAL, the risk associated with the absence of
PAL (as well as with the use of PAL) can therefore vary considerably. The nature and
magnitude of the risk where no PAL is used (scenario 1) still need to be accurately and
clearly communicated so that allergic consumers can make an informed decision. This
requires the use of multiple channels of communication, such as websites, carelines, etc.,
not just the label. However, as PAL is currently voluntary, a product without PAL could also
be one for which no risk assessment has been performed (scenario 2). Such a product
would carry an unquantifiable risk to allergic consumers, rather than one to which an upper
limit has been set.
Risk management, i.e. the minimisation of allergic reactions, is the second role of PAL and it
will be obvious from the foregoing discussion that it can only be discharged successfully if
communication of the PAL message is successful. A pre-condition to success is that the
PAL is observed by (ideally) all at-risk consumers, but this cannot be achieved solely by
setting limits without regard to the implications for the proportion of products that would be
affected. The right balance needs to be struck between the extent to which PAL is used and
any quantitative limits which are set, as already discussed. This is likely to be an iterative
process, and of course, will be influenced by the efficacy of allergen management
procedures. Ward et al. (2010) defined what those levels of protection could mean in
practice. Foods not bearing PAL as a result of risk assessment exercise, although not
specifically designed for people with allergies, would not provoke adverse reactions in the
vast majority of allergic individuals. Allergen management of cross-contact control would be
well-managed. The allergen may be analytically detectable, but the amount is below the
action level (based on reference dose).
5. Implementation of PAL: towards more consistent
allergen risk communication
As mentioned in the previous section, PAL has two closely linked roles: risk management
and risk communication. Allergens should be managed to avoid their unintentional presence
in products. As discussed above, FoodDrinkEurope has developed Guidance on Food
Allergen Management for Food Manufacturers in order to minimize the unintentional
presence of allergens in products and manage the risk deriving from this presence.
Following completion of the risk assessment and elimination or reduction of the risks where
possible through risk management, a decision on whether or not PAL is appropriate then
needs to be made.
Effective and clear risk communication is crucial to ensure that PAL fully plays its role in
protecting allergic consumers. As said before, the inconsistent and, in some cases, unclear
use of PAL, has reduced consumers’ trust and confidence in these warnings and weakened
its role in informing allergic and intolerant consumers. PAL should be communicated to
consumers in a clear, meaningful, consistent way, in order to enable them to correctly
understand the risk and make informed choices when purchasing foods for them and their
families.
This section covers some aspects that, in FoodDrinkEurope’s view, can help moving towards
more consistent risk communication.
Which substances and products should be covered by PAL?
It is important that the scope of PAL is clarified in order to ensure consistent information to
consumers across the EU. In this respect, it is the industry’s understanding that PAL should
cover any substance or product causing allergies or intolerances which is listed in
Annex II to the FIC Regulation.
Should PAL be clearly distinguished from the allergen information given in the list of
ingredients?
As explained, the FIC Regulation requires substances or products causing allergies or
intolerances used in the manufacture or preparation of a food and still present in the finished
product, even if in an altered form, to be indicated in the list of ingredients. It follows that
labelling should clearly distinguish between the allergen information provided in the
list of ingredients and precautionary allergen labelling for those allergens which may
be present.
Expression and presentation of PAL
A consistent approach to the expression and presentation of PAL would facilitate consumers’
understanding and thus ensure the protection of allergic and intolerant consumers.
i) Wording for precautionary allergen statement
There is no legally prescribed wording for PAL. However, the FIC Regulation requires food
information to be given in such a way as not to be misleading, ambiguous or confusing for
the consumer. On the contrary, food information must be accurate, clear and easy to
understand for the consumer.
FoodDrinkEurope considers that the statement used for PAL should take into account these
general requirements and also be brief, simple, and easily translatable into the different EU
languages.
The preferred single harmonised statement for precautionary allergen labelling
recommended by FoodDrinkEurope is: “may contain [allergen]”. “May contain [allergen]” is a
well-known indication for consumers as it has been widely used for many years now. In
order to allow operators to progressively adapt their labels to this single statement, a
sufficiently long transition period is a prerequisite.
ii) Location of wording
With the entry into force of the FIC Regulation, consumers will get used to check the list of
ingredients to see whether the food contains substances or products causing allergies or
intolerances. Therefore, when feasible, the PAL statement should be placed in close
proximity to the ingredient list and be followed by a list of the products or substances for
which an appreciable risk of cross-contamination exists based on quantitative risk
assessment.
In the absence of a list of ingredients, the FIC Regulation states that the indication of the
allergens shall comprise the word “contains” followed by the name of the substance or
product listed in Annex II. In such a case, it is recommended to place the PAL statement in
close proximity to the “contains” statement.
The Regulation foresees a derogation from the obligation to provide information on
allergens, when the name of the food clearly refers to the substance or product concerned
(e.g. milk, butter). In such a case, if an appreciable risk of allergenic cross-contamination
exists, PAL is still recommended. It is up to the operator to decide where to place the PAL
statement on pack, provided that this is clearly legible and visible for consumers.
The above considerations should also be valid in cases where foods are marketed by means
of distance selling.
iii) Other legibility aspects
Other legibility aspects could also be addressed in order to enhance the effectiveness of
PAL:
• General principles: in line with the general requirements applying to mandatory food
information, precautionary allergen information should be easily visible, clearly
legible and, where appropriate, indelible. It shall not in any way be hidden,
obscured, detracted from or interrupted by any other written or pictorial matter or any
other intervening material.
• Font size: The FIC Regulation requires a minimum font size for mandatory allergen
information. However, the use of a minimum font-size is not obligatory for PAL.
Notwithstanding this, the rules applying to mandatory food information with
regard to the minimum font size should also be followed for PAL statements.
The use of a font size bigger than the one chosen for the list of ingredients is not
recommended, as this may induce allergic/intolerant consumers to focus more on the
PAL statement (which informs about the possible presence of allergens) than on the
list of ingredients (which informs about the certain presence of allergens).
• Emphasis: in case of mandatory allergen information, the FIC Regulation requires the
name of the substance or product causing allergies or intolerances to be emphasised
in the list of ingredients, for example by means of the font, style or background
colour. However, there is no legal obligation to emphasise the substances or
products within a PAL statement (which, as mentioned above, should be clearly
distinguished from the list of ingredients). Nevertheless, operators can voluntarily
choose to emphasise the substances or products causing allergies or
intolerances to underline that there is a risk that these are present in the food
(e.g. ‘may contain: milk’).
The above considerations on legibility should also be valid in cases where foods are
marketed by means of distance selling.
PAL for non-prepacked foods
The FIC Regulation requires information on substances or products causing allergies or
intolerances used in the manufacture or preparation of a food and still present in the finished
product, even if in an altered form, to be given also for non-prepacked foods.
Member States may adopt national measures concerning the means through which
mandatory allergen information is to be made available for non-prepacked foods.
Many Member States have already adopted such rules, often allowing for mandatory
allergen information to be given orally provided that certain conditions – which vary
depending on the country – are fulfilled.
Regard should be taken to these national rules, which might also refer to PAL.
It should be noted that PAL for non-prepacked foods would in most cases require
consideration of other factors (handling environment, etc.) which would require development
of further guidance.
Summary Recommendations on PAL
FoodDrinkEurope would like to see a defined framework for the application of PAL which
meets the requirements of Article 36(2) of the FIC Regulation, namely:
• Clear: a single statement with a single meaning, easy to translate into EU languages, i.e.
“may contain [allergen]”
• PAL should not be misleading: it should only be applied where a defined, appreciable
risk has been identified, including (where it is relevant and possible) through a
quantitative risk assessment.
• PAL should be applied based on transparent quantitative limits derived using the most
up to date, relevant, peer-reviewed, and robust scientific data.
• When Quantitative Risk Assessment (QRA) relies on analytical data, it should be noted
that sampling procedures and analytical methods have limitations regarding sensitivity,
specificity, and accuracy. The applicability of analytical data as an input into QRA
requires harmonization.
• In addition, consumers need to know that products have been through a risk assessment
and that the presence or absence of PAL is a consequence of that process.
References
Allen KJ, Remington BC, Baumert JL, Crevel RW, Houben GF, Brooke-Taylor S et al.
Allergen reference doses for precautionary labelling (VITAL 2.0): clinical implications. J
Allergy Clin Immunol 2014; 133(1):156-64.
Allen KJ, Turner PJ, Pawankar R, Taylor SL, Sicherer S, Lack G, Rosario N, Ebisawa M,
Wong G, Mills ENC, Beyer K, Fiocchi A, Sampson HA. Precautionary labelling of foods for
allergen content: are we ready for a global framework. World Allergy Organisation Journal
2014; 7:10-14.
Barnett J, Muncer K, Leftwich J, Shepherd R, Raats MM, Gowland MH et al. Using 'may
contain' labelling to inform food choice: a qualitative study of nut allergic consumers. BMC
Public Health 2011; 11:734.
Blom WM, Michelsen-Huisman AD, van Os-Medendorp H, van Duijn G, de Zeeuw-Brouwer
ML, Versluis A, Castenmiller JJM, Noteborn HPJM, Kruizinga AG, Knulst AC, Houben
GF.2018.
Accidental food allergy reactions: Products and undeclared ingredients. J Allergy Clin
Immunol. 142(3):865-875.
Blom WM, Remington BC, Baumert JL, Bucchini L, Crépet A, Crevel RWR, Madsen CB,
Taylor SL, Houben GF, Kruizinga AG. Sensitivity analysis to derive a food consumption point
estimate for deterministic food allergy risk assessment. Food Chem Toxicol. 2019
Mar;125:413-421. doi: 10.1016/j.fct.2019.01.025. Epub 2019 Jan 25.
Cochrane SA, Gowland MH, Sheffield D, Crevel RW. Characteristics and purchasing
behaviours of food-allergic consumers and those who buy food for them in Great Britain. Clin
Transl Allergy 2013; 3(1):31.
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; 67:262-276.
Dua S, Ruiz-Garcia M, Bond S, Durham SR, Kimber I, Mills C, Roberts G, Skypala I, Wason
J, Ewan P, Boyle R, Clark A.J. Effect of sleep deprivation and exercise on reaction threshold
in adults with peanut allergy: A randomized controlled study. Allergy Clin Immunol. 2019
Dec;144(6):1584-1594.e2.
Dunngalvin A, Chun-Han C, Crevel R, Grimshaw K, Poms R, Schnadt S, Taylor, S, Turner
P, Allen K, Austin M, Baka A, Baumert J, Baumgartner S, Beyer K, Bucchini L, Fernandez-
Rivas M, Grinter K, Houben G, Hourihane J, Kenna F, Kruizinga A, Lack, G, Madsen C, Mills
ENC, Papadopoulos N, Alldrick A, Regent L, Sherlock R, Wal J-M, Roberts G. Allergy 2015,
70, 1039-1051.
DunnGalvin A, Roberts G, Schnadt S, Astley S, Austin M, Blom WM, Baumert J, Chan CH,
Crevel RWR, Grimshaw KEC, Kruizinga AG, Regent L, Taylor S, Walker M, Mills ENC.
Evidence-based approaches to the application of precautionary allergen labelling: Report
from two iFAAM workshops. Clin Exp Allergy. 2019a Sep;49(9):1191-1200.
DunnGalvin A, Roberts G, Regent L, Austin M, Kenna F, Schnadt S, Sanchez-Sanz A,
Hernandez P, Hjorth B, Fernandez-Rivas M, Taylor S, Baumert J, Sheikh A, Astley S, Crevel
R, Mills C. Understanding how consumers with food allergies make decisions based on
precautionary labelling. Clin Exp Allergy. 2019 Nov;49(11):1446-1454.
EFSA NDA Panel (EFSA Panel on Dietetic Products, Nutrition and Allergies), 2014.
Scientific Opinion on the evaluation of allergenic foods and food ingredients for labelling
purposes. EFSA Journal 2014; 12(11):3894, 286 pp. doi:10.2903/j.efsa.2014.3894.
Hefle SL Furlong TJ, Niemann L, Lemon-Mule H, Sicherer S, Taylor SL. Consumer attitudes
and risks associated with packaged foods having advisory labelling regarding the presence
of peanuts. J Allergy Clin Immunol 2007; 120(1):171-176
Holzhauser T, Johnson P, Hindley JP, O'Connor G, Chan CH, Costa J, Fæste CK, Hirst BJ,
Lambertini F, Miani M, Robert MC, Röder M, Ronsmans S, Bugyi Z, Tömösközi S, Flanagan
SD. Are current analytical methods suitable to verify VITAL® 2.0/3.0 allergen reference
doses for EU allergens in foods? Food Chem Toxicol. 2020 Nov;145:111709.
Houben GF, Baumert JL, Blom WM, Kruizinga AG, Meima MY, Remington BC, Wheeler
MW, Westerhout J, Taylor SL.Houben GF, et al. Full range of population Eliciting Dose
values for 14 priority allergenic foods and recommendations for use in risk characterization.
Food Chem Toxicol. 2020 Nov 7;146:111831
Hourihane, J. O., et al., 2017. Peanut Allergen Threshold Study (PATS): Novel single-dose
oral food challenge study to validate eliciting doses in children with peanut allergy. J Allergy
Clin Immunol. 139, 1583-1590.
Jackson L, Al-Taher F, Moorman M, DeVries J, Tippett R, Swanson K, Fu T-J, Salter R,
Dunaif G, Estes S, Albillos S, Gendel S. Cleaning and Other Control and Validation
Strategies To Prevent Allergen Cross-Contact in Food-Processing Operations. Journal of
Food Protection 2008, 71, 445-458
Madsen CB, van den Dungen MW, Cochrane S, Houben GF, Knibb RC, Knulst AC,
Ronsmans S, Yarham RAR, Schnadt S, Turner PJ, Baumert J, Cavandoli E, Chan CH,
Warner A, Crevel RWR. Can we define a level of protection for allergic consumers that
everyone can accept? Regul Toxicol Pharmacol. 2020 Nov;117:104751.
Michelsen-Huisman AD, van Os-Medendorp H, Blom WM, Versluis A, Castenmiller JJM,
Noteborn HPJM, Kruizinga AG, Houben GF, Knulst AC. 2018. Accidental allergic reactions
in food allergy: Causes related to products and patient's management. Allergy. 2018
Dec;73(12):2377-2381.
Remington BC, Westerhout J, Meima MY, Blom WM, Kruizinga AG, Wheeler MW, Taylor
SL, Houben GF, Baumert JL. Updated population minimal eliciting dose distributions for use
in risk assessment of 14 priority food allergens.Food Chem Toxicol. 2020 May;139:111259.
Rimbaud L, Heraud F, La Vieille S, Leblanc J-C, and Crepet A. Quantitative Risk
Assessment Relating to Adventitious Presence of Allergens in Food: A Probabilistic Model
Applied to Peanut in Chocolate. Risk Analysis 2010: 30; 7-19.
Röder M, Baltruweit I, Gruyters H, Ibach A, Mücke I, Matissek R, Vieths S, Holzhauser T.
Allergen sanitation in the food industry: a systematic industrial scale approach to reduce
hazelnut cross-contamination of cookies. Journal of Food Protection 2010, 73, 1671-1679.
Soon, J.M. and Manning, L. 2017. “May Contain” Allergen Statements: Facilitating or
Frustrating Consumers? J Consum Policy, 40: 447-472.
Rolinck-Werninghaus C, Niggemann B, Grabenhenrich L, Wahn U, Beyer K. Outcome of
oral food challenges in children in relation to symptom-eliciting allergen dose and allergen-
specific IgE. Allergy 2012; 67(7):951-957.
SciCom, 2017. Advies 24-2017. Referentie dosissen voor de allergenen die zijn opgenomen
in bijlage II van de Verordening (EU) nr. 1169/2011 van 25 oktober 2011. Wetenschappelijk
Comit´e van het Federaal Agentschap voor de Veiligheid van de Voedselketen.
Stone WE, Yeung JM. Chapter 6: Principles and practice for allergen management and
control in processing. In Allergen Management in the Food Industry (2010) Boye JI and
Godefroy SB (Editors). John Wiley and Sons, Inc. Hoboken, NJ, USA.
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.
Turner PJ, Skypala IJ, Fox AT. Advice provided by health professionals regarding
precautionary allergen labelling. Pediatr Allergy Immunol 2014; 25:290–292.
Turner PJ, Allen KJ, Mehr S, Campbell DE. Knowledge, practice, and views on
precautionary allergen labeling for the management of patients with IgE-mediated food
allergy-a survey of Australasian and UK healthcare professionals. Allergy Clin Immunol Pract
2015; 4:165–167
Waiblinger H.-U. & Schulze G. 2018. Action Levels for Food Allergens: An Approach for
Official Food Control in Germany. J AOAC Int, 101(1): 1-6.
Ward R, Crevel R., Bell I, Khandke N, Ramsay C, Paine S. A vision for allergen
management best practice in the food industry. Trends Food Sci Tech 2010; 21:619-25.
Westerhout J, Baumert JL, Blom WM, Allen KJ, Ballmer-Weber B, Crevel RWR, Dubois AEJ,
Fernández-Rivas M, Greenhawt MJ, O’B Hourihane J, Koplin JJ, Kruizinga AG, Le T-M,
Sampson HA, Shreffler WG, Turner PJ, Taylor SL, Houben GF, Remington BC. 2019.
Deriving individual threshold doses from clinical food challenge data for population risk
assessment of food allergens. Journal of Allergy and Clinical Immunology
Wheeler MW, Westerhout J, Baumert JL, Remington BC. Bayesian Stacked Parametric
Survival with Frailty Components and Interval Censored Failure Times. August 2019.
http://arxiv.org/abs/1908.11334.
Wheeler MW, Westerhout J, Baumert JL, Remington BC. Bayesian Stacked Parametric
Survival with Frailty Components and Interval-Censored Failure Times: An Application to
Food Allergy Risk. Risk Anal. 2020 Oct 16. doi: 10.1111/risa.13585. Online ahead of print.
Zhu J, Pouillot R, Kwegyir-Afful E.K., Luccioli S, Gendel S.M. A retrospective analysis of
allergic reaction severities and minimal eliciting doses for peanut, milk, egg, and soy oral
food challenges. Food Chem Tox 80 (2015): 92-100
***
Annex I
FoodDrinkEurope Statement on Precautionary Allergen Labelling
2
• Allergens are common constituents of consumer products with valuable functional
and/or nutritional attributes which can cause adverse, even life-threatening, reactions in
susceptible individuals. The unintended presence of small amounts of certain allergens
which are not part of a product’s formulation as a result of manufacturing and other
operations (and which are therefore not labelled) can pose a risk to allergic consumers.
• Progressively over the last decades, the food industry has made significant efforts in
reducing the unintended exposure of allergic consumers to major allergens. In
particular, FoodDrinkEurope has developed and published comprehensive Guidance on
Allergen Management for Foods for practical use by operators, which encourages a shift
from the current hazard-based approach to a risk-based approach.
• While current practices in the management of major allergens have increased the safety
of food products for allergic consumers, the lack of an agreed approach to quantitative
risk assessment has led to divergent standards applied by different manufacturers, as
well as divergent approaches by enforcement authorities across Europe. The
subsequent increase in precautionary allergen labelling, in turn, has resulted in reduced
amounts of foods that allergic/intolerant consumers can choose from.
• FoodDrinkEurope considers that precautionary labelling has an important role to play in
protecting allergic consumers, but in order to fulfill that role, it needs to be applied
consistently, in a circumspect manner and in accordance with defined and agreed
principles. FoodDrinkEurope therefore supports a risk-based approach to major allergen
management and the application of precautionary ‘may contain’ labelling.
• Precautionary labelling should only be used where a thorough risk assessment
demonstrates that there is a real risk of a significant but unavoidable amount of allergen
in the consumed product due to cross-contact within the ingredient supply chain or from
manufacturing operations. Although we recognize that following a risk-based approach
may cause reactions in a very small proportion of susceptible individuals, this approach
will minimize risk to consumers with food allergies, while maximizing their food choices.
• To clarify when precautionary allergen labelling applies and to further facilitate its
optimal use for consumers, FoodDrinkEurope supports the development of EU-wide
harmonised approach with transparent and applicable limits based on latest scientific
evidence and guidance on appropriate forms of wording for labelling statements. This
will allow industry to consistently apply precautionary labelling and clearly communicate
the allergen status of a food.
• Given the globalisation of the food chain FoodDrinkEurope recognises that the
development of a harmonised global risk-based approach would be optimal and also
supports activities aiming to achieve this.
2
For ease of reference, precautionary allergen labelling referred to in this statement refers to the labelling of
substances and/or products causing allergies or intolerances in accordance with Regulation (EU) 1169/2011 on
the provision of food information to consumers.