Article Text

Download PDFPDF

Incidence of severe and fatal reactions to foods
  1. J O’B Hourihane1,
  2. D Reading2,
  3. P Smith3,
  4. G Lack4,
  5. D Hill5,
  6. A Muñoz-Furlong6,
  7. S A Bock7
  1. 1Wellcome Trust Clinical Research Facility, Southampton University Hospitals NHS Trust, Southampton, UK
  2. 2The Anaphylaxis Campaign, PO Box 275, Farnborough, Hampshire, UK
  3. 3Brisbane, Australia
  4. 4St Mary’s Hospital, London, UK
  5. 5Department of Allergy, Children’s Allergy Centre, Royal Children’s Hospital, Parkville, VIC 3052, Australia
  6. 6The Food Allergy & Anaphylaxis Network, 10400 Eaton Place, Fairfax, VA 22030, USA
  7. 7National Jewish Medical and Research Center, Department of Paediatrics, University of Colorado Health Sciences Center, Denver, CO, USA
  1. A Colver8,
  2. C Macdougall9,
  3. A Cant10
  1. 8Northumbria Health Care Trust and University of Newcastle upon Tyne, Donald Court House, 13 Walker Terrace, Gateshead NE8 1EB, UK
  2. 9Newcastle General Hospital, Westgate Road, Newcastle upon Tyne NE4 6BE, UK
  3. 10Paediatric Immunology and Infectious Diseases Unit, Newcastle General Hospital, UK

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Although the article by Macdougall et al1 regarding the incidence of severe and fatal reactions to food would be seem to be reassuring, we would like to express some concerns and raise some questions about the data presented. The first question is whether the ascertainment of cases is really as complete as the authors suggest. We acknowledge that the UK medical system may allow better reporting and access to mortality data than that of the US. However, the records acquired as described seem to represent the same underreporting issues as those in the US. Is it really unlikely that the BPSU misses a significant number of cases? Based upon a well characterised population in Olmstead county Minnesota and extrapolating the data to a US population of 280 million, it may be estimated that there are 200 deaths from anaphylaxis reactions to food each year.2

A paper published in 2001, described methodology in which a National Registry had been established and was well publicised to US allergists 3 Very few reports were made by allergists and none by other physicians. No cases were initially reported by physicians who conduct research in food allergy. Nearly all the cases were ascertained from the press. These news articles appeared in local newspapers and were not reported in media with a large regional or national circulation. In an earlier effort to account for all cases of food anaphylaxis, only in Colorado, a significantly higher number of cases were reported from rural regions as compared to metropolitan areas strongly suggesting either misdiagnosis or inaccurate recording of cases in the emergency department log of busy hospitals.4

A second concern is the reporting of cases only up to age 15. In the paper mentioned above, of 32 fatalities 10 occurred in youngsters up to age 15.3 An additional 10 occurred in adolescents aged 16 to 19. Why did MacDougall et al not include all adolescents?

A third question must always be raised when fatal food anaphylaxis is studied. Is it not possible that cases of fatal asthma were actually initiated by unidentified allergic reactions to food? All authors in this field are likely to agree that the ultimate cause of death may be irreversible airway obstruction, and all would agree that poorly controlled asthma increases the risk of fatal anaphylactic reactions to food, but we would suggest that the trigger responsible for individual asthma fatalities is not always determined. What about fatalities that never reach the emergency department and are misclassified on death certificates as asthma fatalities? Individuals that die at home and are classified as asthma deaths are unlikely to be further investigated in either the US or the UK.

Fourthly, the authors’ definition of severity seems incomplete. Individuals with severe food reactions who self administer epinephrine often do not go to hospital, are less likely to have reactions that require hospitalisation or cause death, and often they do not report these reactions to their physicians unless specifically queried. Some survive the reaction without treatment, become convinced that they know to avoid a specific food, and never tell their physician. We could argue about the possible progression of these episodes to near fatal or fatal reactions, but the point to be made is that they are frequently under reported. The fifth issue concerns the safe administration of epinephrine. We disagree about the risk to children of the administration of a single dose of epinephrine as opposed to withholding that dose. We have no disagreement about aggressive treatment of asthma concurrently, and in fact we think that point should be emphasised. However families reading this commentary may become more fearful, than they currently are, about administering epinephrine. We know that epinephrine is not always life saving even when administered in a timely fashion,5 however withholding it surely must increase the risk of calamity. Over dosage certainly may occur, but it seems more likely that an overdose would be administered by medically trained personnel than by parents. The over prescription of epinephrine is a debatable issue, however it seems a small price to pay, with a low risk, in order to save even one young life.

Finally, we are very concerned that families will interpret this paper to mean that death from food allergy is very unlikely, and therefore they may relax their vigilance. If families of younger children become less concerned when their children become adolescents it may to difficult to institute a good prevention education program. This is exactly the opposite of the goal of education programs in the US (The Food Allergy and Anaphylaxis Network, and UK (The Anaphylaxis Campaign) aimed at making individuals with food allergy and the general population more aware of the problem and the potential for mortality. It is truly unfortunate that we cannot accurately identify all of the individuals who die during allergic reactions to food and use this information to do a better job of preventing these tragedies. We must continue our campaigns of education of medical professionals and the public, and we must be certain that emergency treatment is available when and where it is needed.


Authors’ reply

We thank Bock et al for their interest in our article. We respect their views on the interpretation of the data but it is of course for each reader to come to their own opinion on these. We would like to respond to their comments on the accuracy and valididity of our data.

Did our paper under ascertain deaths? Bock et al base their concerns on our methods of case ascertainment and on comparison with another study. We cannot be certain about this but as the text indicated we used many sources and spoke to many experts in the field. We agree we did not search local newspapers but this would have been almost impossible as few were on CD-ROM in the 1990s. As mentioned, we did search national newspapers and all cases we came across were already known through one of our other sources. Finally, since publication, no-one has told us of a case we appear to have missed.

We specifically studied children up to 15 years because this is the group we were interested in. Many recommendations on risks to children are based on inferences from data covering all ages and we wanted to bring a proper paediatric perspective. Indeed the interpretation Bock et al give to the paper they cite1 is grossly misleading. They suggest extrapolation to a US population would lead to 200 deaths from food each year; yet the paper, in which there is only one death (occurring during exercise), covers all ages and reactions to all allergens, not just food

The issue of whether asthma deaths may have been precipitated by food allergy is an important question which we addressed “If a child’s symptoms are only asthmatic and no allergen is suspected, then there is no means for attributing such reactions to food or for knowing if a causal link exists”. Furthermore, such deaths will never have been reported in surveys of food allergy in other countries or in other age groups. No group has been able to address this question satisfactorily and it is a key area for further research.

We are not sure we agree that children, who have self administered epinephrine, often do not go to hospital. However we do not know the proportion and said as much, excluding this group from our definition of severity.

Finally we agree that education of professionals and the public should continue based on the best data available. This must include those parents whose children are truly at high risk as well as those many parents that think any immediate hypersensitivity reaction to food means their child is at high risk of an allergic death; when in reality the risk, in the absence of asthma, seems very small. Different parents will come to different views about how to proceed faced by a severe but very small risk, just as we all do in many aspects of our lives.