Healthcare professionals’ beliefs and practices regarding food allergy testing for children with eczema

Atopic dermatitis/eczema (hereafter 'eczema') affects around 20% of children in the UK and symptoms can be difficult to manage.1 Parents of children with eczema often request food allergy tests or exclude foods from their diets to manage the symptoms.2 Whilst it is understood that early-onset eczema is associated with food allergy and allergic reactions to foods can exacerbate eczema symptoms,3 evidence that foods cause long-term eczema symptoms is weak.4.

Participants were given information on the first page of the survey to read before deciding whether to take part or not. Consent was implied when participants clicked past the first page and answer the questions on page two. HCPs who diagnose and treat children with eczema were invited, via professional networks and social media, to participate anonymously via a GDPR compliant survey host (www. onlin esurv eys.ac.uk). The survey was open for four weeks, June-July 2020. Quantitative data were analysed using Stata v15, and the chisquare test was used to compare all four levels of response across all four HCP specialties. Formal analysis of free text responses was not performed, but instead responses were grouped under broad themes and quotes selected to illustrate the views of some respondents.
First, HCPs were asked about the use of food allergy tests for children with eczema, with or without a clinical history of allergic reaction to food (see Figure 1). For children with eczema and a history of immediate reaction, more HCPs in allergy (91%) and paediatrics (73%) always requested a test, compared to GP (25%) and dermatology (25%; p < .001). For children with no history of reaction, most HCPs in GP (88%), paediatrics (53%) and dermatology (58%) never requested a test but most HCPs in allergy (77%) requested a test sometimes or more often (p < .001). For children with a history of delayed reaction, HCPs across all specialties, allergy (82%), GP (76%), paediatrics (73%) and dermatology (75%) requested a test sometimes or more often (p = .583). For children with a mixed picture of immediate and delayed reactions, rates of testing were higher amongst HCPs in allergy (91% always or mostly) and paediatrics (90%) compared with dermatology (50%) and GP (31%; p < .001).
Next, HCPs were asked about the use of food allergy tests for children with different severities of eczema when there is no history of a clinical reaction to food ( Figure 2). Allergy HCPs were more likely than other HCPs to test for food allergy in children with clear eczema (82% never compared with 97%-100%, p < .017). HCPs in allergy and paediatrics used food allergy tests more often than HCPs in GP or dermatology for children with mild (never: allergy 59% and paediatric 76%, vs dermatology 82% and GP 98%; p < .001), moderate (always or mostly: allergy 27% and paediatrics 10% vs dermatology 0% and GP 3.1%; p = .006), severe (always or mostly: allergy 45% and paediatrics 40% vs dermatology 33% and GP 14%; p = .001) and very severe (always or mostly: allergy 45% and paediatrics 53% vs dermatology 42% and GP 25%; p = .010) eczema.
Finally, HCPs were asked about the use of food allergy tests in cases of faltering growth, parent requests, early-onset eczema, difficult to treat eczema and family history of atopy. Regarding faltering growth, many HCPs never requested a food allergy test in GP (60%), paediatrics (43%) and dermatology (42%), whereas most HCPs in allergy did sometimes or more often (67%; p = .048). There was no difference between specialties when considering parent requests in the decision to request an allergy test (sometimes or never GP 95%, paediatrics 90%, allergy I try to avoid it. Difficult to interpret, but there is mounting parental pressure fuelled by the media to do these kinds of tests.
I rarely request these tests but parents ask for them a lot.
To our knowledge, this is the first study to evaluate HCPs' beliefs and practices regarding food allergy testing and eczema. We found opinions varied by scenario and speciality, with HCPs in allergy and paediatrics more likely overall to request a food allergy test than HCPs in GP or dermatology; and HCPs in allergy and paediatrics varying their practice more than the other groups according to the clinical scenario.
The online survey allowed us to reach healthcare professionals across the UK, but our findings are limited by the number of, and manner by which, the sample were recruited: half of the sample comprised cli-

ACK N OWLED G EM ENTS
None.

CO N FLI C T S O F I NTE R E S T
RJB has received honoraria for participating in advisory boards for ALK-Abello, DBV technologies and Prota therapeutics, who research or manufacture treatments for people with food allergy.

AUTH O R CO NTR I B UTI O N S
MJR conceived and led on the study; AG designed the online version of the survey and led on the analysis; RJB and SM helped with the overall study design, delivery and interpretation of the findings. AG wrote the first draft of the research letter with assistance from MJR and input from RB and SM. All authors reviewed, commented and approved the final manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
Requests for data will be considered by the corresponding author.