The epidemiology of infantile eczema

I found that infants with eczema were at greatly increased risk of an IgE-mediated food allergy to peanut, egg white or sesame. One in five infants with eczema were allergic to one or more of these foods at age 14 months, compared to one in twenty-five infants without eczema. Increasing eczema severity and a younger age at eczema onset were risk factors for food allergy among infants with eczema, as were increasing socio-economic status and East Asian ethnicity. Pet ownership, on the other hand, decreased the odds of food allergy among infants with eczema by 40-50%. I found no evidence for a role of skin barrier function at non-lesional skin at age 14 months as a mechanism by which sensitisation to food allergens may occur, however I found some evidence of skin barrier dysfunction at eczema lesion sites.

In my PhD, I investigated three aspects of eczema epidemiology in infancy: the current prevalence of eczema; potential risk factors for eczema; and the progression from eczema to IgE sensitisation, IgE-mediated food allergy, and atopic asthma.

I found no evidence for a role in the causation of eczema for several factors associated with microbial exposure or the maternal or infant diet. One exception was maternal consumption of peanut during breastfeeding, which was associated with a decreased risk of infantile eczema (OR 0.8, 95% CI 0.6, 1.0). Maternal consumption of probiotics was associated with an increased risk of infantile eczema, however this was explained by a trend towards probiotics consumption among mothers who have eczema, asthma and/or allergic rhinitis. Infant consumption of antibiotics- a factor that has been associated with an increased risk of eczema in multiple studies- was not plausibly a causative factor for most cases of eczema, as eczema onset usually preceded antibiotic use. Furthermore, the crude association between infant antibiotic use and eczema was largely explained by the increased need for treatment of skin infections among patients with eczema.

There has been intense interest in the potential role of vitamin D deficiency as a risk factor for food allergy and asthma. My PhD research does not, however, support a role for vitamin D in causing a child's eczema to occur: the prevalence of eczema did not vary by season of birth; serum vitamin D levels were comparable at age 14 months between infants with and without eczema; mother's with a history of vitamin D deficiency were no more or less likely to have a child with eczema; and there was no association found between maternal sun exposure and eczema incidence. My results suggest that sun exposure and vitamin D may have a role in modifying eczema severity, with increased maternal sun protection in pregnancy associated with a reduced eczema severity score at age 14 months.

As well as being a highly prevalent condition of early childhood, I found that eczema was associated with long-lasting sequelae. Infants and young children with eczema were highly likely to have atopic asthma that was current in middle age, particularly when allergic rhinitis was also experienced. There was no association between eczema and non-atopic asthma. Whether or not there is a causal link between eczema and other atopic disease, there is great need to understand the natural history of eczema in the atopic march and to counsel patients accordingly.

Eczema is a common disease of early childhood which can have major effects on a child's development and quality of life. The prevalence of eczema has been rising over the past few decades and is particularly high in Australia. The rise in eczema prevalence is occurring too rapidly to be explained by genetic factors alone, however environmental factors that can explain the rise in eczema have yet to be fully elucidated. An important aspect of eczema epidemiology is the progression from eczema to other allergic disease. The prevalence of food allergy among infants with eczema from the general community is not known, whilst the idea of a progression from eczema to asthma in an 'atopic march' is controversial. It is unclear whether there is a causative role for eczema or skin barrier dysfunction in the development of sensitisation and IgE-driven disease.

My results show that eczema remains a common childhood disease: approximately one in three infants (38.5%) are affected. Certain aspects of the HealthNuts Study methodology lend support to this being one of the most valid estimates of infantile eczema prevalence for an urban population: a high response rate, broad sampling frame, and information about eczema collected even among families who declined participation in the study.

The strongest risk factors for eczema that I could identify related to socio-demographic factors, rather than easily-modifiable factors. Male sex (OR 1.4, 95% CI 1.2, 1.6), maternal eczema (OR 2.1, 95% CI 1.7, 2.7) and paternal eczema (OR 1.8, 95% CI 1.4, 2.5) were strong predictors of infantile eczema risk. Additionally, infants of East Asian ethnicity were found to be at greater risk of eczema than infants whose parents were born in Australia (a largely Caucasian population) (OR 1.6, 95% CI 1.3, 1.9 for each parent born in East Asia). This was in spite of the East Asian parent group having a lower prevalence of allergy compared to Australian-born parents: hence there is evidence for gene-environment interactions which trigger an eczema susceptibility among East Asian migrants to Melbourne.

To address my research questions, I drew on data from two large, Australian observational studies. The HealthNuts Study has recruited approximately 5300 twelve-month-old infants (as of mid-2011) from the community in Melbourne, Australia, with the aim of measuring the current prevalence and finding risk factors for allergic disease in infancy. This thesis examines data for the first 3405 participants. All infants were examined for eczema, underwent a skin prick test to common food allergens (cow's milk, hen's egg, peanut, sesame and/or shellfish) to measure sensitisation (2mm wheal or greater). Uniquely, IgE-mediated food allergy was confirmed among sensitised infants using standardised oral food challenge protocols, under medical supervision. Parents of participating infants provided information about their child's history of eczema and potential risk factors for allergic disease. To explore the association between eczema and atopic asthma, I drew on data from the Tasmanian Longitudinal Health Study. This study is one of the longest running asthma studies in the world: over eight thousand Tasmanian children were first sampled at age seven years, in 1968, and over five thousand were followed-up most recently in 2004, at age approximately 44 years. A sub-sample of the 2004 participants underwent sensitisation testing to a panel of common inhalant allergens, which permitted phenotyping of adult asthma into atopic and non-atopic forms.