Adjusted for male sex, child exposure to active smoking and household resident number.

* Test for trend across exposure categories: - refers to the linear increase in risk if none; one, or two synthetic items were used ( for frequent wheeze, p=0.004)

+ Reference category

In the cross-sectional sample, there was a significant trend for increased likelihood of frequent wheeze with increased amounts of synthetic bedding. Synthetic bedding was associated with increased frequent when compared to feather bedding (aRR=1.7 (1.1-2.4) but not other non-feather bedding (aRR=1.2 (0.7, 2.1). However, feather bedding was very uncommon in infancy, so could not explain the adverse effect of infant synthetic pillow use.