Issue 22 / 17 June 2013

AUSTRALIAN immunologists have questioned the use of salicylate elimination diets for children, with a group of paediatric immunologists writing in the MJA that the diets are not supported by evidence and may be harmful. (1)

In a literature review and a retrospective review of case notes from the authors’ own practices looking at children who had been on a low salicylate diet, the authors said they could find no peer-reviewed evidence supporting the use of the diet in conditions for which it is prescribed, such as eczema and behavioural disturbances.

They also identified 31 adverse events from the case notes — from failure to thrive (13 cases) and food aversion (six cases) to eating disorders (four cases) and alopecia (two cases) — in the 66 children for whom this information was available.

“The use of low salicylate diets in children is not supported by current evidence or by expert opinion”, wrote Dr Paul Gray, of the department of immunology and infectious diseases at Sydney Children’s Hospital, and his coauthors. “While our retrospective case note review is insufficient to prove any risk associated with the diets, it is concerning that harm may occur when children and adolescents are placed on such restrictive diets, particularly if they stay on them for long periods.”

However, in a commentary accompanying the research, Professor Robert Loblay, of the department of clinical immunology at Sydney’s Royal Prince Alfred Hospital, and coauthors said that “with suitable guidance” salicylate elimination diets based on n-of-1 testing were safe and could improve quality of life. (2)

“The results from peer-reviewed publication of group data may not be transferrable to determine what dietary modifications are appropriate in individual cases; the diagnostic elimination–challenge process we use is essentially a single-patient (n-of-1) trial method which is generally considered to provide strong evidence in guiding clinical practice”, the authors wrote, adding that intake of nutrients by children on a low chemical diet was usually at or about the recommended daily amount.

Professor Geoff Cleghorn, professor of paediatrics at the University of Queensland, told MJA InSight that while the literature review had found no evidence of benefit from low salicylate diets, the researchers had drawn a long bow in linking the diets to harm.

“The authors’ analysis would suggest that the children who were on a low salicylate diet did suffer from adverse events, but whether this was because of the low salicylate per se, or because of other things that were being promulgated with the diet, it’s impossible to tell from their evidence”, Professor Cleghorn said.

“It’s very unlikely that low salicylate diets are done in isolation, they are done with a whole bunch of other food removals.”

While Professor Cleghorn said he was unconvinced that the salicylate elimination diet was harmful, he was equally uncertain of any benefit. He said the commentary did not add to the evidence, presenting only anecdotal reports of benefit.

“I’m not sure that I’m convinced that [a low salicylate diet is] hazardous … But equally, does one need to do it?” he said. “Few things we do come without risk.”

Dr Christine Ziegler, paediatric allergist/clinical immunologist at Flinders Medical Centre, said there had been a “massive surge” in IgE and non-IgE mediated food allergies to key foods including egg, dairy, wheat and soy in infants and young children. She said it was these common foods that most often needed to be targeted with dietary modification.

However, food chemical intolerances, including salicylate intolerances, occurred very infrequently in children aged under 10 years, Dr Ziegler told MJA InSight.

“We don’t tend to use [salicylate elimination] diets, because the kids are presenting with actual food allergies, as opposed to food chemical intolerance”, she said.

Salicylates and other food chemicals were seen across a wide range of different foods. “If salicylates are targeted with dietary elimination, it can cut out a wide range of foods from the diet unnecessarily and can contribute to restricted nutritional range”, Dr Ziegler said.

In a research paper published in the same issue of the MJA, researchers found that parents of children with food allergies were often complacent about precautionary labelling on food, regardless of whether their children had a history of anaphylaxis. (3)

The researchers concluded that the use of fewer precautionary statements and more effective labelling strategies may lead to less consumer complacency.


1. MJA 2013; 198: 600-602
2. MJA 2013; 198: 603
3. MJA 2013; 198: 621-623

7 thoughts on “Salicylate diet claims “unconvincing”

  1. Chris says:

    My take on this diet (having had intolerance to salyicilates and allergic reactions to dairy and gluten is that it is a meams of coping with symptoms only and does not get to the underlying cause. Who knows why, but most of the members of my family have some level of problem with lots of food groups (my brother and I have very limited diets) otherwise we feel very unwell, to the point where we can’t work because we are mentally compromised, physically impaired and generally unwell. I sometimes go back to the elimination diet when I need to give myself a break from sals, as a way of resetting. However, wheat, barley, rye and all dairy are out permanently.

    Having been living with these issues for more than 10 yrs now, I would welcome further advances in the science. Even if the science is lacking, trial and error are what get results in practice with a knowledgeable dietitian – at least that is how it worked for me.

  2. Deborah Doyle says:

    I’ve been on the RPA low salicylate, no amines and no glutamates, no preservatives, additives, colourings or flavourings etc. diet. At first the diet seemed to help my symptoms of red, flushed skin often accompanied by a feeling of sunburn — the burning feeling so intense that it  sometimes hurt me to wear clothes just as it does when truly sunburned. At some point I added dairy back in, specifically milk and butter — I added milk and butter back in within about a year of starting the diet  — succesfully. I was doing a bit better with some reduction in that symptom (but far from total).The sunburned feeling reduced in intensity. There were periods of time, when I was less ‘reactive,’ when I was able to eat things like a bit of cream cheese, even  a few slices of tomato very occasionally, sweet potato etc. But not a lot of variety — no important orange veges. However the general trend that emerged was that I was becoming more reactive to foods. Now my diet has become so restrictive that it can no way be considered healthy yet the one symptom it seemed to help – the sunburned feeling – is stronger than ever now. I believe that in my case harm did occur. My subjective opinon – yes. But, so far, I’ve not seen any  objective scientific evidence that this diet is not harmful in the long term and I’ve looked all over the web for it. I don’t think it’s job of the immunologists that criticise the diet to provide evidence of harm. I think it is the moral responsibility of the people who are selling the diet to provide more objective science based study on its efficacy, safety, or harm. And they are selling it.

  3. Naomi Cook says:

    “…insufficient to prove any risk associated with the diets… harm may occur when children and adolescents are placed on such restrictive diets.”

    So, let me get this right: Risk can’t be proven, yet in the same breath, ‘harm’ is basically listed as a potential risk?

     ‘Harm’ ‘may’ occur at any point in time, to any cohort, sample or group of individuals, children or teenagers.

    How many preschoolers, due to fussy appetites are walking around with undiagnosed nutritional deficiencies, (iron, Vitamin D being the most obvious). What about eating disorders (including anorexia, as the authors have specifically highlighted), they too are an unfortunate but common feature of teenage years in our society. We also know – from evidenced based studies – that they tend to have multi-factorial (often non-food related) causative factors.

    Now, the authors do admit they have a sample bias and no control group and so one might perhaps expect a less dramatized conclusion than the one that, in spite of itself, stretches to predict ‘harm.’

    I do however agree with their final point:

    We would invite any proponents and prescribers of the diet to produce evidence of the efficacy and safety for the disorders.”

    As a Registered Nurse and mother of a food sensitive family who have been on a low salicylate diet for 2.5 years (and are the healthiest we have ever been), I would love to see a secure evidence base for the sole purpose that more may benefit from the dietary changes that have transformed the lives of both myself and my children.

  4. Dr Howard Dengate says:

    Gray and colleagues’ evidence, based on allergy clinic findings, is no better than the science attacked, as they acknowledge: “our cohort has an inherent selection bias and … without a control group”.

    On nutrition, Dumbrell and colleagues concluded that the nutritional quality of the low salicylate diet was superior to that of the normal diet: “With proper dietary counselling, the elimination test diet is safe for use in the treatment of children …”  [1].

    The 9,200 member Food Intolerance Network ( provides independent information and support for families using the RPAH (Royal Prince Alfred Hospital) elimination diet. We always recommend consulting a dietitian and we provide a list of supportive and experienced dietitians (email who can supervise members through the elimination, challenge and reintroduction phases of their diet.

    We strongly support the peer-reviewed publication of evidence regarding the effects of dietary salicylates on health, behaviour and learning and acknowledge that more research needs to be published.

    We have received thousands of reader reports about remarkable improvements due to a low salicylate diet, usually tried as a last resort after numerous treatments recommended by other health professionals have failed.  Some of them can be seen at

    [1] Dumbrell S et al. Is the Australian version of the Feingold diet safe? Med J Aust. 1978;2(12):548, 569-70.

  5. RF Heller says:

    The contributions by Gray and Loblay and colleagues on the topic of salicylate intolerance in the current issue of the MJA well illustrate that ‘absence of evidence does not equate to evidence of absence’ [1]. Neither paper quotes the, admittedly sparse, European peer reviewed literature pointing to a good understanding of the basic biochemistry [2,3], a diagnostic test for and clinical prevalence studies of intolerance to dietary salicylate [3-5]. There is also a wealth of anecdotal evidence about the benefits of a low salicylate diet on gastrointestinal symptoms in the community, with support groups such as the Food Intolerance Network reporting many ‘success stories’ It is unfortunate that there is a lack of high quality clinical and epidemiological research into what may be a common problem and one for which, in the absence of greater understanding of the underlying causes, selective and appropriately implemented dietary manipulation may be really helpful.

    [1] Altman DG & Bland MJ Statistics notes: Absence of evidence is not evidence of absence BMJ 1995;311:485

    [2] Paterson J et al. Is there a role for dietary salicylates in health? Proceedings of the Nutrition Society 2006;65:93-96

    [3] Baenkler H-W. Salicylate intolerance: Pathophysiology, Clinical Spectrum, Diagnosis and Treatment. Dtch Arztebl Int 2008;105:137-142.

    [4] Zopf Y et al. The differential diagnosis of food intolerance. Dtch Arztebl Int 2009;106:357-370.

    [5] Raithel M et al. Significance of salicylate intolerance in diseases of the lower gastrointestinal tract. J Physiol and Pharm 2005;56 (Supp 5):89-102.

  6. Joy Anderson says:

    As a dietitian who uses the RPAH Allergy Unit Elimination Diet in my everyday practice, I can assure you that it does work brilliantly in the majority of cases, in infants (via mother’s breastmilk), in children and in adults as well. However, it needs to be done properly. From what is written above, it sounds like people just do this on their own indefinitely. Saying that it is useless because of this is like throwing the baby out with the bath water. The diet should be supervised by an Accredited Practising Dietitian with experience in food-chemical intolerances and conducted as a test diet, for a limnited period of time – usually only 3-4 weeks in duration. Occasionally it may go longer, but the dietitian ensures that nutrients are adequately compensated for. The challenges are then performed in a timely manner and the diet refined to be liberised as much as possible, while only avoiding the problem foods long term. I have many, many satisfied clients who were fobbed off by other health professionals in the past, because they didn’t ‘believe’ in food-chemical intolerance.

  7. Mark Donohoe says:

    Between 1989 and 1993, two medical colleagues and I tested patients formally in a controlled inpatient environment in a Sydney hospital. We used the RPAH Allergy Clinic’s test capsules at that time, and found salicylate adverse reactions insignificantly different to placebo, at arount 2-3%. I recall we were criticised at that time as lacking scientific rigour because we failed to replicate the well accepted findings of a major Sydney hospital. 

    How the broad promulgation of the restrictive low salicylate diet has persisted for the two decades since without a reasonable evidence base remains a mystery.

    Part of the problem is the time period some patients are put on the restrictive test diet, failing to improve but persisting against all common sense. On occasions beyond a full year, waiting for the symptom-free days after which they can challenge.

    Another problem is the conviction – sometimes bordering on obsession – that results in the maintenance of the restrictive diet despite failure of symptom reduction, and evidence of harm from the dietary restriction as time passes.

    The concept of “n-of-1” testing is certainly valuable in empirical medicine, but eventually may need to be tested by well constructed trials. Otherwise it starts to seem more like an irrefutable and self-evident truth, and the placebo and nocebo effects can become invisible.

Comments are closed.