Discussion

Summary

We present a flowchart to aid the interpretation of dietary history and skin prick test results from children with AD less than two years of age. This tool is based on a consensus exercise with a multi-disciplinary panel of clinicians with expertise in the management of children with AD and/or food allergies. It is underpinned by advice on which 12 or 7 symptoms to consider relevant to possible immediate or delayed allergy respectively and which skin prick test thresholds to regard as negative or sensitised. Consensus was reached for the use of commercial reagents for wheat and soya but not which specific regents to use for milk or egg skin prick tests.

Strengths and limitations

This is the first published attempt to use a formal consensus process to agree what dietary advice should be given to parent of children with AD based a given combination of dietary symptoms and skin prick tests results to four foods, which are commonly allergenic or are suspected by parents as causing their child’s AD symptoms. The exercise benefitted from with perspective of clinicians from different disciplines, working in generalist and specialist care. Panellist engagement was high, frank views were shared and anonymity was retained until the workshop.
A consensus exercise has recently been undertaken to address uncertainties with the detection and management of milk allergy in children under two years of age.[12] Modifications to the Delphi process are not an isolated practice in the health sciences[13] and were necessary for the sake of brevity.[14] The online format allowed the panel to be drawn from across the UK, so findings could reflect practice variations which may exist between regions and populations being served, and support the panel member commitment to the whole process. The short timescale over which the consensus exercise was conducted may have helped recall of previous rounds and maintained engagement.
Nevertheless, the findings will not apply to all young children with AD and opinions outside of the 14 panel members and the UK may be more varied. It is also possible that had all panellists been able to engage with all stages, the findings may been different. Although participants were proficient at contributing and exchanging views online, persuasive arguments may have been made or received differently in a face-to-face meeting. Skin prick tests for milk and egg are more reliable than wheat or soya and these results do not directly correlate with specific IgE levels from blood tests. Although decision-making about food allergy testing and subsequent dietary modifications should be shared and discussed with parents and carers in a clinical consultation, because of its technical nature patients representatives were not involved in this process. However, the findings do begin to address the research priorities identified by patients, carers and clinicians of “What role might food allergy tests play in treating eczema?” and “What is the role of diet in treating eczema: exclusion diets and nutritional supplements?”[15]
The complexity of food allergy in children is such that it can be argued that decision-making cannot be protocolised.[16] Allergy history taking takes time, expertise and can be a subtle art, where through careful conversation, concerns or symptoms can be identified as relevant or reassurance given. This is particularly the case with severe AD of early onset, certain symptom complexes (e.g. co-occurrence of skin and gut symptoms), when multiple foods are implicated and/or there is a prior food allergy diagnosis. Similarly, evidence of impact on the growth and/or micronutrient deficiency trajectory of a child are important considerations.[17] Although treatment plans can be decided and advised with multidisciplinary support, their implementation and maintenance relies on parents/carers understanding and ability to adhere to them for the long-term. In the absence of any benefits outweighing potential harms from routine food allergy to guide dietary advice for the management of AD,[8] parents should still be supported to use topical therapies to treat their child’s skin.
In consensus exercises, the context in which questions are presented is important.[14] The same questions presented in another setting may have elicited different responses. The context for this exercise, where skin prick tests are performed on “all-comers”, departs from usual clinical practice where symptoms are usually assessed first to determine the merit of food allergy tests. It also assumes that for any child who needs it, oral food challenges will be done in a timely manner, something which currently does not happen in the UK, for example. Finally, it also supposes that parents and carers of trial children will be supported to safely exclude or reintroduction of study food(s) as appropriate.

Findings in the context of the literature

Dietary modification by parents of children with AD, seeking to reduce symptoms, is common and often without any healthcare input.[6] Undiagnosed delayed food allergy as the cause for AD symptoms and the use of blood specific IgE or skin prick food allergy tests to guide dietary advice is controversial. Clinical opinions are divided with healthcare professionals but those working in allergy and paediatrics are more likely to request a food allergy test than those working in dermatology or primary care.[7, 18] Mortz et al have published an algorithm (based on “expert opinion and available literature”) on “When and how to evaluate for immediate (IgE‐mediated) food allergy in children with atopic dermatitis”.[19] This similarly recommends that a careful dietary history and an assessment of AD severity is obtained prior to any allergen testing. Consequently, decision making for children with no symptoms, immediate symptoms and/or moderate-severe AD is relatively straightforward. On-going research, based on our findings, will help to address the gap where there are parental concerns in a child with mild AD and/or possible delayed symptoms.[20]
A recent systematic review of 10 RCTs concluded that dietary elimination “may lead to a slight, potentially unimportant improvement in AD severity, pruritus, and sleeplessness” in patients with mild to moderate AD.[8] However, only four of the included studies had interventions based on positive allergy tests, oral food challenges, or both.[21-24] There was limited data on potential harms from such interventions and as the authors point out, any potential benefits must be balanced against the psychosocial impact of food exclusion, developmental and nutritional problems and indeed the risk of developing IgE-mediated food allergy.[25]

Implications for research, policymakers or clinicians

Until further research clarifies the relationship between food allergy and AD, and the role of food allergy tests, use of IgE food allergy tests in clinical settings should continue be confined to help confirm or refute the diagnosis of immediate-type allergies, following an allergy focused history.[18] Clinicians need to be mindful that parental concerns about food allergy and modifications to the diets of children with AD is under-recognised and a dissonance can exist between parental and clinicians views.[26] Regardless of opinions about food allergy testing, clinicians should seek to identify and address these issues as they can be a barrier to effective use of topical treatments and have consequences for nutrition and diet quality.
Although food allergy testing may be sought by parents of children with AD to guide dietary exclusions, the findings from this exercise can also guide introduction, reintroduction and inclusion of foods where parents are unsure based on symptoms, or have misplaced concerns which are addressed by a negative skin prick test result.