![]()
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The changing epidemiology of food
allergy—implications for New Zealand
Christine Crooks, Rohan
Ameratunga, Greg Simmons, Penny Jorgensen,
Clare Wall, Maia Brewerton, Jan Sinclair, Richard Steele, Shanthi
Ameratunga
Food allergy backgroundFood allergy (FA) is a common adverse reaction to food
mediated by the immune system. The condition can result in a wide variety of
clinical manifestations that involve IgE or non-IgE mediated mechanisms, or
sometimes both (e.g. eosinophilic oesophagitis). In contrast, food intolerances
can be considered adverse non-allergic responses to foods. This is best
illustrated by lactase deficiency, where cow’s milk can cause gut symptoms
as a result of undigested lactose.1
The gastrointestinal tract is important not only in the
absorption of nutrients but also in the protection against microbial invasion
through appropriate immune responses. Oral tolerance develops when there is
down-regulation of the immune response to non-harmful antigenic
substances.2
Allergic reactions to food are largely thought to be due to
genetic factors and can occur when oral tolerance is
impaired.2 The severity of FA can vary from
trivial abdominal discomfort to death from anaphylaxis. FA accounts for a
significant proportion of anaphylaxis and severe allergic reactions in
children.3–5 Much less is known about the
pattern of FA in older adults.6
Most allergic reactions to food originate in childhood and
few treatment options have the potential to alter their natural
course.7 Both genetic and environmental factors
are likely to be involved in the pathogenesis of FA, but the interactions are
complex with considerable gaps in knowledge.7
Studies of identical twins indicate high concordance pointing to genetic
predisposition, but inter-related environmental factors such as feeding
practices are also likely to be
important.8
Many children who develop allergy to food may outgrow their
allergies.9–12 For example, in a study of
118 children with cow’s milk allergy (CMA), 42 (35.6%) with non-IgE
mediated CMA were found to be free of their allergy by the age of 5
years.12 However 13 children (15% of the
cohort) had persistent CMA at the age of 8.6 years. Thus, contrary to previous
perceptions, CMA may persist into late childhood in a significant minority of
patients.12
Accurate diagnosis and appropriate management of food
allergy is critical. Failure to identify the offending food allergen(s)
correctly may place the person at risk of recurrent anaphylaxis. On the other
hand, inappropriate and unsupervised dietary elimination can increase the risk
of nutritional deficiencies. This was recently illustrated by the case of a
14-month-old child with CMA who presented with rickets due to poorly managed
dietary exclusions, resulting in deficiencies in Vitamin D, dietary calcium, and
phosphate causing impaired bone mineralisation and
growth.13
Therefore, FA sufferers are doubly challenged to find a diet
that will not result in an adverse reaction, but which is nutritionally
balanced, in order to maintain good
health.14
In addition to the clinical symptoms, severe FA can pose a
significant socioeconomic burden to families as diet and lifestyle may be
adversely affected. Unfortunately, some of the commonest allergenic foods are
also of high nutritional value (e.g. milk, eggs, and peanuts) relative to their
cost. The financial burden of substituting alternative dietary options can be
substantial.13 FA can also limit social
activities and school attendance in children, and young adults may have
restricted career options.15
FA can generate considerable anxiety in affected families
and the community, and some patients suffer post-traumatic stress disorder after
a severe episode of anaphylaxis. Indeed, a recent study suggested that having a
child with severe food allergy had the same adverse impact on the quality of
life as having a child with Type 1 diabetes.15
Prevalence of food allergy: problems in ascertainmentFA is acknowledged to be a significant public health problem
in developed countries, but there are major gaps in knowledge regarding the
population burden of the condition as highlighted in a recent meta-analysis
funded by the European Commission (part of the EuroPrevall research project).
Of the 934 papers that investigated the prevalence of
allergy from January 1990 to December 2005,16
only 54 met the authors’ criteria for inclusion, and only 19 included the
double-blind placebo-controlled food challenge (DBPCFC) method of diagnosis
(considered to be the gold standard
method14,16–18).
The articles that met the review criteria were classified by
the diagnostic methods used; self-reported FA (SRFA), specific IgE (IgE cut-off
levels varied between studies), skin prick test, a combination of SRFA and IgE
or skin prick test, and SRFA and DBPCFC. The authors found widely varying
estimates of prevalence depending on the method used. For example, the
prevalence of CMA ranged from 1.2 to 17%, egg allergy from 0.2 to 7%, and peanut
allergy from 0 to 2%.16 Not surprisingly,
studies based only on self-reported FA have tended to provide the highest
estimates of prevalence.16
Inconsistent study designThe variability in reported results was subsequently
highlighted by Keil, who critically reviewed published studies on the
epidemiology of FA between October 2005 and January
2007.19 Only six published studies were
identified that met the review criteria for study design, recruitment process,
assessment methods of FA outcomes, and
interpretation.19 These included two
well-designed birth cohort studies from the Isle of Wight in the United Kingdom
(UK).
The first of these studies found that the prevalence
estimates of FA (based on clinical history, skin prick testing, open food
challenges and DBPCFC) in the first year of life varied between 2.2% and 5.5%,
which were considerably lower than the estimates based on parental reports
(between 5.5% and 14.2%).17
In the second Isle of Wight study, FA prevalence was
determined in a cohort of 6 year-olds. Adverse reactions to food were reported
by 11.8% of the cohort. This is higher than the prevalence confirmed by clinical
history, skin prick tests, and open food challenge (2.2%) and DBPCFC
(1.6%).20
The propensity to over-estimate adverse reactions to food
based on self-reported symptoms has also been observed among teenagers. For
example, Pereira et al found that in contrast to the prevalence of self-reported
symptoms among 11 year olds (11.6%) and 15 year olds (12.4%), only 2.2% had a
diagnosis confirmed by food challenges.21
Similar rates may also be perceived by adults; in a study of 1483 adult subjects
in the Netherlands, 12.4% reported FA but only 0.8% were confirmed as FA by
DBPCFC.22
Inconsistent study design has been identified as a problem
in the Global Allergy and Asthma European Network
(GA2LEN) review of 18 on-going European birth
cohort studies.23 It is hoped that through the
co-operation of participating research teams that some data may be pooled and
common analyses used for endpoints such as the natural history of
FA.23
Laboratory factors may influence the diagnosis of FA,
including food-specific IgE cut-off values.16
Limitations also exist due to a lack of standardisation of the skin test
allergens used to assess sensitisation.19,23 A
survey of the participating research teams in the
GA2LEN study found that different standard
panels of allergens are used across
Europe.23
There may be variability in physician diagnosis which
further complicates the perception of increasing FA
prevalence.9 Responses to a questionnaire sent
to 7000 United States physicians indicated that non-allergist physicians
diagnosed FA at a higher rate than allergy specialists. This is the first study
to report differences in FA diagnosis between physicians. This survey suggests
there may be a need for further training in this area.
The establishment of accurate data on FA prevalence is
problematic because most estimates of prevalence are based on methods other than
a comprehensive approach including symptoms, allergy testing, and the gold
standard DBPCFC.
It is difficult to predict clinical reactivity to an
allergen based purely on the measurement of IgE antibodies in serum and mast
cell reactions by SPT.24 Positive results from
either method, especially with low levels of food-specific IgE, do not mean that
an allergic reaction is inevitable on consumption of the
food.24 The significance of other methods such
as kinesiology, hair testing, and ‘electroaccupuncture according to
Voll’ (EAV) are unknown.
One approach that could address the problems described is to
recruit a large unselected birth cohort, where regular clinical assessments and
allergy testing is undertaken. Children suspected of having FA then undergo
DBPCFC. Such studies are, however, limited by expense and logistical
complexities, and may not provide time-dependent-information (such as changing
population demographics and changing dietary practices), and do not provide
information about adults with FA. Furthermore, there may be ethical concerns
with undertaking DBPCFC in children.
Alternatively, monitoring the change in patterns of food
allergy (using the same research tools used to detect changes in prevalence of
peanut allergy18 and
asthma25) would be a simpler and less expensive
approach, and could provide relevant, useful information.
Is the incidence of food allergy increasing?The extent to which the burden of FA has changed over time
is contested. While some studies do not support evidence of increasing incidence
of FA,14,16,19 the well-designed Isle of Wight
birth cohort studies showed an increase in prevalence of peanut allergy (from
0.5 to 1.0%) and peanut sensitisation (from 1.1 to 3.3%) from 1989–1994 to
1994–1996.26
By using the same research instrument (nationwide,
cross-sectional, random telephone survey with a standardised questionnaire),
Sicherer et al found self-reported peanut allergy doubled in children less than
5 years of age from 1997 to 2002 in the US.18
Clinical reports of children with food-related anaphylaxis have also reportedly
increased in Australia.27,28 The explanations
for these observed increases were not investigated in these studies.
Food allergy in non-European populationsFA reactions appear to occur at a higher rate in Asian
children in Westernised countries.29,30 A
cohort study produced important results on self-reported wheeze in European
children and south Asian children born in the
UK.29 Parental reports of food and drink
triggered-wheeze were significantly higher in the south Asian children compared
to the Europeans and doubled over a 5-year
period.29 While this was a clinically
significant finding due to the impact on health services, the reasons for these
differences were not reported.
Moreover, FA may be more common in Asian countries than
previously suspected.8 In Singaporean children
(identifying with Chinese, Malay, Indian, and Eurasian ethnic groups) the most
common foods causing allergies were peanuts, shellfish, and
egg.8
In this study of FA, which was determined by SPT and
questionnaire, peanut allergy was seen in a third of children. While Singaporean
children develop shellfish allergy at an older age than peanut or egg allergies,
it is the second most common allergen.8
This finding for Singaporean children was considered to be
different to paediatric norms in the United States and Western Europe where the
major allergenic foods are milk, egg, and
peanuts.8
For Singaporean and Hong Kong adults, shellfish allergy is
the major cause of anaphylaxis in patients reporting to emergency
departments.8 The high incidence of peanut
allergy was also in contrast to previous results. It had been thought that the
low incidence of peanut allergy in Chinese populations was due to different
processing methods. FA appears to be of concern in Asian countries, and the
authors recommended that large-scale epidemiological studies be carried out
there.8
These studies highlight the changing epidemiology of food
allergy and underscore the importance of environmental factors in food
allergy.
Food allergy in New Zealand: what is known?Recent publications have included limited data on FA
prevalence in New Zealand.16,31 In the first
study, information was from the early 1990s and was part of the European
Community Respiratory Health Study. Data were collected from 1148 New Zealanders
by a brief questionnaire about respiratory health, which included four questions
regarding dietary intake.32
From this study 11.4% of subjects reported illness from
food, although it is unclear if the FA diagnosis was supported by allergy tests.
In another paper the prevalence of CMA amongst New Zealanders was reported to be
about 11% based on child and parental reported data from a cohort of 155
children aged 3 to 10 years in Dunedin.31
The burden and characteristics of FA in New Zealand are
likely to have changed over time in view of the demographic changes (e.g.
substantial increases in the proportions of Asian and other immigrants in the
last two decades) and changes in the diversity and production of foods available
to consumers.
Several species of shellfish are endemic to New
Zealand.33 Allergic reactions to shellfish and
molluscs may be highly cross-reactive, which is thought to be due to the
conserved nature of amino-acid sequences in the allergenic protein tropomyosin
across species.34
The relationship between FA and New Zealand’s unique
species of shellfish may also require further investigation. Infant feeding
patterns and genetic predisposition are also likely to vary in different ethnic
communities. While equivalent data for FA are not available, Pacific children
who migrate to New Zealand from countries such as Samoa and Tonga have been
noted to experience an increased incidence of asthma compared to children in
their home countries.35
The incidence of FA in Māori is unknown. In order to
fully understand the aetiology of FA and initiate primary prevention and
treatment strategies in New Zealand, it is important to identify any disparities
between Māori and non-Māori.
Māori have on average the worst health status of any
ethnicity in New Zealand.36 The Māori
population has a higher growth rate compared to Europeans, and between 2006 and
2021 the Māori population is expected to grow by 20% compared to only 5% in
Europeans.37 Therefore the paucity of
information on FA in Māori limits our ability to predict the future disease
burden and plan the most appropriate delivery and access to health resources
over the coming decades.
Overseas research has identified FA as a risk factor for
life-threatening asthma attacks in children.38
Māori experience greater morbidity associated with asthma and Māori
are twice as likely to be hospitalised for their asthma as
non-Māori.36 In asthma and other atopic
diseases, FA may have an important role in the health disparities we see between
Māori and non-Māori.
Lower socioeconomic status is a well documented predictor of
poor health outcomes and a barrier to care; it is well documented that
Māori are more likely to live in more deprived areas than
non-Māori.36 Therefore, the collection and
analysis of up-to-date, accurate epidemiological data and the development of
specific health strategies must remain an ongoing priority in order to tackle
these inequalities.
Consequences of the limited data on food allergy in New ZealandAnecdotal evidence shows that the public hospital system in
New Zealand experiences substantial constraints in responding to the needs of
those affected with FA.
Children with the potential for outgrowing FA need to be
reviewed regularly so that they are not confined to an unnecessarily restricted
diet, while adults with FA also need to be reviewed particularly after
experiencing a severe reaction.
With no national guidelines for the provision of allergy
services there is an ad hoc approach by district health boards (DHBs)
including the availability of specialists and purchase of laboratory tests.
Few New Zealand hospitals currently offer specialist allergy
services for adults, and services for children are particularly limited from a
national perspective. Indeed, many patients with complex allergic disorders have
to fly long distances for diagnosis and treatment, often at their own expense.
Similar shortages of allergy specialists and primary health care services are
also seen in the UK13,39 and in
Australia.27
Access to laboratory tests in some parts of the country is
also restricted. In Wellington, for example, patients of private specialists are
required to pay for laboratory tests.
Better epidemiological data may assist health boards in
prioritising the need for allergy/immunology services and ensure robust
coordination and continuity of care across primary, secondary, and (where
necessary) tertiary care services. Tertiary hospital-based multidisciplinary
teams would ideally include specialist allergists, nurses, dietitians, and
facilities for food challenges and immunotherapy.
A better understanding of the epidemiology of FA may also
identify the gaps in established protocols and school services. The ad
hoc approach to severe food allergy in New Zealand’s schools may
place some students at increased risk of reactions while attending school.
Currently, adrenaline auto-injector devices—the
primary treatment for acute anaphylaxis in the community—are not publicly
funded in New Zealand. Ensuring an up-to-date supply of this life-saving
medication (e.g. EpiPen®—the commonly
used formulation—has a shelf-life of 12–16 months) can place an
unacceptably high financial burden on families with severe FA. Better data on
the prevalence of severe allergies may assist funding these devices.
Compliance with food-allergen labelling regulations is a
significant issue for the food industry including manufacturers, the food
service, and hospitality sectors. Mistakes can be costly for both the industry
and consumers. Better data on the prevalence of food allergy would assist the
food industry both in compliance and in producing food suitable for people with
FA.
Studies of FA can provide useful information that is
relevant to the national economy. New Zealand is heavily dependent on
agricultural exports and the development of new foods. Recently, the kiwi fruit
gold variety was identified as being as allergenic as the green variety in
Europe.40
The inadvertent development of highly allergenic foods may
inflict damage to New Zealand’s international reputation as an exporter of
high quality foods. On-going FA studies have the potential to identify these
foods at an early stage, or ensure appropriate processing and labelling to
mitigate the risk to vulnerable individuals.
While many advances have been made in the last two decades,
much has yet to be learned about FA. More targeted research elucidating the
burden, barriers to effective treatment, and related factors in New Zealand is
necessary to ensure better services and support for all children and adults with
food allergy, and improved safety for those at risk of severe reactions.
Competing interests: Unrestricted
educational grants have been provided by Allergy New Zealand, ASCIA, Australian
Laboratory Services, Nutricia, and William and Lois Manchester Trust for this
research.
Author information: Christine Crooks,
Research Fellow, Department Virology and Immunology, LabPlus, Auckland Hospital,
Auckland; Rohan Ameratunga, Associate Professor, Adult and Paediatric Allergy
Specialist, Department Virology and Immunology, LabPlus, Auckland Hospital,
Auckland; Greg Simmons, Medical Officer of Health & Public Health Physician,
Auckland Regional Public Health Service, Greenlane Clinical Centre, Auckland;
Penny Jorgensen, Chief Executive Officer, Allergy New Zealand, Auckland; Clare
Wall, Senior Lecturer, Department Human Nutrition, University of Auckland; Maia
Brewerton, Immunology Registrar, Department of Respiratory Medicine, Wellington
Hospital, Wellington; Jan Sinclair, Paediatric Immunologist, Department of
Paediatrics, Starship Hospital, Auckland; Richard Steele, Clinical Immunologist,
Wellington Hospital, Wellington; Shanthi Ameratunga, Associate Professor,
Section of Epidemiology & Biostatistics, School of Population Health,
University of Auckland, Auckland
Acknowledgements: The authors thank ADHB
for their continuing support; Allergy New Zealand, ASCIA, Australian Laboratory
Services, Nutricia, and William and Lois Manchester Trust for unrestricted
educational grants; Associate Professor Scott Sicherer and Prof Hugh Sampson for
their helpful comments; and Drs Vincent Crump and Allen Liang for support.
Correspondence: Associate Professor Rohan
Ameratunga, Department of Virology and Immunology, LabPlus, Auckland Hospital,
Park Rd, Grafton, Auckland, New Zealand. Fax: +64 (0)9 3072826; email: rohana@adhb.govt.nz
References:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |