The epidemiology of childhood eczema: I. A population based study of associations.

Information on whether they thought their child had ever had eczema was obtained from the mothers of 12,555 children in a national cohort of five-year-olds born in 1970. This question was part of a multiple battery of questions concerning the medical, social, environmental and behavioural background of the child. These data were linked to the information that had been collected on the cohort at birth, and a profile of characteristics of the children with reported eczema was produced. A large proportion (46/135) of associations were statistically significant at the 1% level. The major associations were with socio-economic indicators and characteristics of parental health behaviour, with the most advantaged socio-economic groups and those with more positive health behaviour having increased rates of reported eczema. The patterns of associations form an interesting profile of the backgrounds of children reported to have had eczema. Identification of these factors was necessary before more advanced statistical techniques were employed to investigate which of these variables predominate when they are considered simultaneously, and to generate hypotheses as to which factors may be causally associated with the disorder.